Communication method and apparatus

Abstract

A communication method and apparatus, which can rationally formulate a corresponding transmission policy, and avoid the waste of communication resources. In the method, a first node can acquire first information, and send the first information to a second node, wherein the first information comprises model transfer performance indicator information of a first model, which model transfer performance indicator information is used for indicating the model transfer performance of the first model. In this way, the second node can determine the model transfer performance of the first model, and rationally plan communication resources required for model transfer, thereby avoiding the waste of communication resources, and achieving relatively high model gains.

Description

Communication method and apparatus

[0001] The present application claims priority from the Chinese patent application No. 202411093840.7 filed on August 8, 2024, and entitled "Communication method and apparatus", the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0002] The present application relates to the field of communication, in particular to a communication method and apparatus. BACKGROUND

[0003] In a wireless communication network, as the diversification of service requirements and the enhancement of network functions, service implementation, network planning, configuration, and resource scheduling also become increasingly complex. For example, service implementation, network planning, configuration, and resource scheduling may involve modulation, coding, transmitters, receivers, multi-antenna technology, or positioning technology in a wireless communication system. Among them, the network device can implement, for example, signal modulation and demodulation, information encoding and decoding, channel state information (CSI) feedback, beam management (BM), or mobility management, etc. by performing relevant operations.

[0004] Currently, in the process of implementing the above-mentioned schemes through a wireless communication network, data transmission is usually involved between communication devices, for example, the network device needs to transmit the required model data to the terminal, and the data transmission not only affects the actual execution effect of the terminal, but also causes communication overhead. However, in the actual process, the network device in the related technology is difficult to reasonably formulate a corresponding transmission strategy, thereby affecting the normal execution of the relevant operation or causing waste of communication resources. SUMMARY

[0005] In order to solve the above technical problems, the embodiments of the present application provide a communication method and apparatus, which can reasonably formulate a corresponding transmission strategy to avoid waste of communication resources.

[0006] In a first aspect, a communication method is provided. The method can be performed by a first node, or by a component of the first node, such as a processor, a chip, or a chip system of the first node, or by a logic module or software that can implement all or part of the functions of the first node. Hereinafter, the method is taken as an example to be performed by the first node. The communication method comprises: obtaining first information; the first information comprises model transmission performance indicator information of a first model; wherein the first model is used to enable artificial intelligence (AI) / machine learning (ML) functions; the model transmission performance indicator information is used to represent the model transmission performance of the first model; and sending the first information to a second node.

[0007] Currently, the network side is difficult to evaluate the expected performance of the terminal side after model delivery, which causes the model delivery strategy formulated by the network device to fail to guarantee high model expected revenue. For example, the network device is limited by communication resources and can only deliver part of the model, but the delivered model does not achieve good model revenue, which causes the model delivery strategy formulated by the network device to fail to achieve good model revenue, and at the same time, occupies a large amount of communication resources, affecting the normal delivery of other models. Based on this, the first node can send the related information of the model delivery performance indicator of the first model to the second node, so that the second node can fully consider the model delivery performance of the first model (also referred to as the performance-price ratio of model delivery, or also referred to as the revenue value of model delivery, or also referred to as the target performance of model delivery, or also referred to as the stability of model delivery, or also referred to as the robustness of model delivery) when formulating the model delivery strategy. Therefore, the second node can reasonably plan the communication resources required for model delivery based on the related information of the model delivery performance indicator, avoid wasting communication resources, and at the same time achieve higher revenue of the model.

[0008] In a second aspect, a communication method is provided. The method can be executed by a second node, or by a component of the second node, such as a processor, a chip, or a chip system of the second node, or by a logic module or software that can realize all or part of the function of the second node. The following takes the method executed by the second node as an example. The communication method includes: receiving first information from a first node; the first information includes model delivery performance indicator information of a first model; the first model is used to enable an artificial intelligence (AI) / machine learning (ML) function; the model delivery performance indicator information is used to represent the model delivery performance of the first model; and determining a model delivery strategy according to the first information.

[0009] The technical effects of the second aspect can refer to the first aspect, which will not be repeated here.

[0010] In combination with the first aspect or the second aspect, in a possible design, the model delivery performance indicator information includes at least one model delivery performance indicator, and the at least one model delivery performance indicator includes at least one model delivery performance indicator value and / or at least one first value range. Since the model delivery performance of the model is affected by changes in the channel environment, communication configuration adjustment, or external physical interference, in this case, the first node can report the model delivery performance related information of the model in the form of a value or a value range according to the current situation, so that the second node can consider the influence of various factors when determining the model delivery strategy, and ensure the normal use of the model.

[0011] With reference to the first aspect or the second aspect, in a possible design, the at least one model delivery performance indicator value includes at least two model delivery performance indicator values. The at least one first value range includes at least two first value ranges. In this way, the first information can include multiple values / value ranges of the model delivery performance indicator, so that the second node has more selection space when determining the model delivery strategy.

[0012] With reference to the first aspect or the second aspect, in a possible design, the model delivery performance indicator information includes mathematical statistics used to represent the model delivery performance of the first model. That is, the model delivery performance indicator information can be determined by statistically analyzing the model delivery performance, so that the model delivery performance indicator information can more accurately or robustly reflect the model delivery performance of the first model.

[0013] With reference to the first aspect or the second aspect, in a possible design, the model delivery performance indicator information includes mathematical statistics used to represent the model delivery performance indicator of the first model.

[0014] With reference to the first aspect or the second aspect, in a possible design, the first information includes at least one model delivery performance indicator information; one or more model delivery performance indicator information in the at least one model delivery performance indicator information has a corresponding data feature or data feature identifier.

[0015] In an embodiment of this application, a feature can be used to represent a feature possessed by a signal, a channel, or data (for example, data that can be used to train the first model or data that can be used for inference / processing of the first model) associated with the signal or the channel, due to device self-configuration or environmental factors, for example, one or more of node deployment, antenna form, and transceiver waveform, in a communication process of a target node (for example, the first node or the second node) or in a process of executing the first model. Different nodes can have different features, and the difference can affect training and / or inference of the first model. Therefore, in an embodiment of this application, the model delivery performance indicator information can be bound to the data feature or the data feature identifier, so that the first node and the second node can consider the impact of device self-configuration or environmental factors when performing related operations.

[0016] With reference to the first aspect or the second aspect, in a possible design, the at least one model delivery performance indicator information is in one-to-one correspondence with the at least one data feature or data feature identifier.

[0017] With reference to the first aspect or the second aspect, in a possible design, the at least one model delivery performance indicator is in one-to-one correspondence with the at least one data feature or data feature identifier.

[0018] With reference to the first aspect or the second aspect above, in a possible design of the first aspect or the second aspect, the first model corresponds to at least one model delivery performance indicator information, and the at least one model delivery performance indicator information includes first model delivery performance indicator information, the first model delivery performance indicator information corresponding to the first data feature or the first data feature identifier.

[0019] With reference to the first aspect or the second aspect above, in a possible design of the first aspect or the second aspect, the at least one model delivery performance indicator information further includes second model delivery performance indicator information, the second model delivery performance indicator information corresponding to the second data feature or the second data feature identifier.

[0020] With reference to the first aspect or the second aspect above, in a possible design of the first aspect or the second aspect, the model delivery performance indicator information is determined according to state information and / or model delivery information; the state information includes first active state information and / or first inactive state information.

[0021] With reference to the first aspect or the second aspect above, in a possible design of the first aspect or the second aspect, the model delivery information includes first model delivery quantization information; the first model delivery quantization information includes model delivery times and / or model delivery time.

[0022] With reference to the first aspect or the second aspect above, in a possible design of the first aspect or the second aspect, the first active state information corresponds to an active state of the first model, and the active state includes at least one of the following: a state in which performance of the first model reaches or exceeds a first target performance indicator; an activated state of the first model; or a supportable / applicable / available state of the first model; and / or the first inactive state information corresponds to an inactive state of the first model, and the inactive state includes at least one of the following: a state in which performance of the first model does not reach or exceed a second target performance indicator; a non-activated state of the first model; or a non-supportable / non-applicable / non-available state of the first model.

[0023] With reference to the first aspect or the second aspect above, in a possible design of the first aspect or the second aspect, the performance of the first model reaching or exceeding the first target performance indicator includes that the performance of the first model reaches or exceeds a value of the first target performance indicator, or is within a value range of the first target performance indicator; and the performance of the first model not reaching or exceeding the second target performance indicator includes that the performance of the first model does not reach or exceed a value of the second target performance indicator, or is not within a value range of the second target performance indicator.

[0024] Since the performance of the first model will usually change due to various factors when actually used, in the embodiments of the present application, the performance of the first model can be divided into multiple intervals according to the value or value range, for example, the performance interval of the value of the performance of the first model reaching or exceeding the first target performance index is the interval corresponding to the effective state, or the performance interval of the value range of the first target performance index is the interval corresponding to the effective state; the performance interval of the value of the performance of the first model not reaching or not exceeding the second target performance index is the interval corresponding to the invalid state, or the performance interval of the value range of the second target performance index is the interval corresponding to the invalid state. In this way, the current state of the first model can be determined based on the interval in which the performance of the first model is located.

[0025] In combination with the first aspect or the second aspect, in a possible design, the first effective state information includes first effective quantification information of the first model, the first effective quantification information including effective time information and / or effective frequency information, wherein the effective time information includes a length of time during which the first model is in the effective state, and the effective frequency information includes a number of times during which the first model is in the effective state; and / or, the first invalid state information includes first invalid quantification information of the first model, the first invalid quantification information including invalid time information and / or invalid frequency information, wherein the invalid time information includes a length of time during which the first model is in the invalid state, and the invalid frequency information includes a number of times during which the first model is in the invalid state. In this way, the effective state information and the invalid state information can be quantified according to different dimensions, such as a time dimension and a statistical frequency dimension, so as to facilitate evaluation of the model transfer performance index information of the first model.

[0026] In combination with the first aspect or the second aspect, in a possible design, the model transfer performance index information includes at least one of the following: first proportion information, first difference information, and first probability statistical information.

[0027] In combination with the first aspect or the second aspect, in a possible design, the first proportion information includes: a proportion of the first effective quantification information and the model transfer information, or a proportion of the first invalid quantification information and the model transfer information, or a proportion of the model transfer information and the first effective quantification information, or a proportion of the model transfer information and the first invalid quantification information.

[0028] With reference to the first aspect or the second aspect, in a possible design, the first effective state information is first effective state information determined in at least one first time period; and / or, the first ineffective state information is first ineffective state information determined in at least one second time period. Since the model transfer performance of the first model can be affected by changes in a channel environment, adjustment of a communication configuration, or external physical interference, which is manifested as a change in the model transfer performance over time, the first effective state information and the first ineffective state information can be counted in a time-segmented manner in embodiments of this application, to further determine the model transfer performance indicator information of the first model at different times.

[0029] With reference to the first aspect or the second aspect, in a possible design, the at least one first time period includes at least one of: at least one time period in which performance monitoring is performed on the first model; at least one time period in which the first model is in an activated state; at least one time period in which the first model is in a supportable / suitable / available state; at least one time period indicated by the second node; or, at least one time period determined by the first node; and / or, the at least one second time period includes at least one of: at least one time period in which performance monitoring is performed on the first model; at least one time period in which the first model is in an activated state; at least one time period in which the first model is in a supportable / suitable / available state; at least one time period indicated by the second node; or, at least one time period determined by the first node.

[0030] With reference to the first aspect or the second aspect, in a possible design, the at least one first time period is included in at least one monitoring time window or at least one activated time window or at least one suitable time window or at least one available time window, and the at least one monitoring time window or the at least one activated time window or the at least one suitable time window or the at least one available time window is earlier than the first time unit in which the model transfer performance indicator is sent; and / or, the at least one second time period is included in at least one monitoring time window or at least one activated time window or at least one suitable time window or at least one available time window, and the at least one monitoring time window or the at least one activated time window or the at least one suitable time window or the at least one available time window is earlier than the second time unit in which the model transfer performance indicator is sent.

[0031] With the first or second aspect above, in a possible design, the first time unit is not located in the monitoring time window, or the first time unit is not located in the activation time window, or the first time unit is not located in the applicable time window, or the first time unit is not located in the available time window; and / or, the second time unit is not located in the monitoring time window, or the second time unit is not located in the activation time window, or the second time unit is not located in the applicable time window, or the second time unit is not located in the available time window.

[0032] In a third aspect, a communication apparatus is provided, which can implement various methods. The communication apparatus includes modules, units, or means corresponding to the methods, and the modules, units, or means can be implemented by hardware, software, or by a combination of hardware and software. The hardware or software includes one or more modules or units corresponding to the functions.

[0033] In some possible designs, the communication apparatus can include a processing module and a transceiver module. The processing module can be configured to perform the processing functions in any of the above aspects and any possible implementation manner thereof. The transceiver module can include a receiving module and a sending module, which are configured to perform the receiving function and the sending function in any of the above aspects and any possible implementation manner thereof.

[0034] In some possible designs, the transceiver module can be composed of a transceiver circuit, a transceiver, a transceiver, or a communication interface.

[0035] In a fourth aspect, a communication apparatus is provided, which includes a processor and a memory. The memory is configured to store computer instructions, and the processor is configured to execute the instructions, so that the communication apparatus performs the method in any of the above aspects and any possible design thereof.

[0036] In a fifth aspect, a communication apparatus is provided, which includes a processor and a communication interface. The communication interface is configured to communicate with modules outside the communication apparatus. The processor is configured to execute computer programs or instructions, so that the communication apparatus performs the method in any of the above aspects and any possible design thereof.

[0037] In a sixth aspect, a communication apparatus is provided, which includes at least one processor. The processor is configured to execute computer programs or instructions stored in a memory, so that the communication apparatus performs the method in any of the above aspects and any possible design thereof. The memory can be coupled with the processor, or can be independent of the processor.

[0038] In a seventh aspect, a communication apparatus (for example, the communication apparatus can be a chip or a chip system) is provided, which includes a processor configured to perform the functions in any of the above aspects and any possible design thereof.

[0039] In some possible design, the communication apparatus includes a memory, which is configured to store necessary program instructions and data.

[0040] In some possible design, the apparatus is a chip system, which can be composed of a chip or include a chip and other discrete devices.

[0041] The communication apparatus in the third aspect to the seventh aspect can be the first node in the first aspect, or an apparatus included in the first node, such as a chip or a chip system; or the communication apparatus can be the second node in the second aspect, or an apparatus included in the second node, such as a chip or a chip system.

[0042] The eighth aspect provides a communication apparatus, which can be the first node, or a module or unit (for example, a chip, or a chip system, or a circuit) corresponding to the method / operation / step / action described in the first aspect executed in the first node, or a module or unit capable of being used with the first node; or the communication apparatus can be the second node, or a module or unit (for example, a chip, or a chip system, or a circuit) corresponding to the method / operation / step / action described in the second aspect executed in the second node, or a module or unit capable of being used with the second node.

[0043] It can be understood that, when the communication apparatus in any of the third aspect to the eighth aspect is a chip, the sending action / function of the communication apparatus can be understood as outputting information, and the receiving action / function of the communication apparatus can be understood as inputting information.

[0044] The ninth aspect provides a computer-readable storage medium, which stores a computer program or instructions, and when the computer program or instructions are executed on the communication apparatus, the communication apparatus can perform the method in any of the aspects and any possible design thereof.

[0045] The tenth aspect provides a computer program product including instructions, and when the computer program product is executed on the communication apparatus, the communication apparatus can perform the method in any of the aspects and any possible design thereof.

[0046] The technical effects brought by any design in the third aspect to the tenth aspect can refer to the technical effects brought by different design in the first aspect or the second aspect, which will not be repeated here. BRIEF DESCRIPTION OF DRAWINGS

[0047] FIG. 1 is a structural schematic diagram of a communication system provided in the present application;

[0048] Fig. 2 is a schematic diagram of a structure of another communication system provided in the present application;

[0049] Fig. 3 is a schematic diagram of a structure of an O-RAN system provided in the present application;

[0050] Fig. 4 is a schematic diagram of a flow of a model transfer method provided in the present application;

[0051] Fig. 5 is a schematic diagram of a flow of another model transfer method provided in the present application;

[0052] Fig. 6 is a schematic diagram of a flow of a communication method provided in the present application;

[0053] Fig. 7 is a schematic diagram of a scenario of a model transfer performance provided in the present application;

[0054] Fig. 8 is a schematic diagram of another scenario of a model transfer performance provided in the present application;

[0055] Fig. 9 is a schematic diagram of a structure of a time window provided in the present application;

[0056] Fig. 10 is a schematic diagram of another structure of a time window provided in the present application;

[0057] Fig. 11 is a schematic diagram of a flow of a communication method provided in the present application;

[0058] Figs. 12-14 are schematic diagrams of structures of communication apparatuses provided in the present application. DETAILED DESCRIPTION

[0059] In the description of the present application, unless otherwise specified, “ / ” represents that the objects before and after the “ / ” are in an “or” relationship, for example, A / B can represent A or B; “and / or” in the present application is only a description of the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent: A exists alone, A and B exist together, and B exists alone, where A and B can be singular or plural.

[0060] In the description of the present application, unless otherwise specified, “multiple” means two or more than two. “At least one of the following” or the like means any combination of the items, including any combination of single item or multiple items. For example, at least one of a, b, or c can represent: a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.

[0061] In addition, in order to facilitate clear description of the technical solutions of the embodiments of the present application, in the embodiments of the present application, the terms "first", "second", etc. are used to distinguish the same or similar items or items with basically the same function and role. Those skilled in the art can understand that the terms "first", "second", etc. do not limit the quantity and execution order, and the terms "first", "second", etc. also do not mean that they are necessarily different.

[0062] In the embodiments of the present application, the words "exemplary" or "for example" are used to mean serving as an example, instance, or illustration, any implementation or design solution described as "exemplary" or "for example" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other implementations or design solutions. Rather, the use of the words "exemplary" or "for example" is intended to present the relevant concept in a specific manner, facilitating understanding.

[0063] It can be understood that the "embodiments" mentioned throughout the specification mean that the specific features, structures or characteristics related to the embodiments are included in at least one embodiment of the present application. Therefore, the various embodiments throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It can be understood that in various embodiments of the present application, the size of the sequence number of each process does not mean the execution order, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0064] It can be understood that in the present application, "when" and "if" refer to the corresponding processing under certain objective circumstances, not the time limit, and do not require judgment actions when implementing, nor mean that there are other limitations.

[0065] It can be understood that some optional features in the embodiments of the present application can be implemented independently in some scenarios without relying on other features, such as the scheme currently based on, to solve the corresponding technical problems and achieve the corresponding effects. Also, in some scenarios, they can be combined with other features according to demand. Correspondingly, the devices given in the embodiments of the present application can also implement these features or functions, which will not be described here.

[0066] In the present application, except for special description, the same or similar parts of each embodiment can be mutually referred. In the various embodiments of the present application, if there is no special description and no logical conflict, the terms and / or descriptions of different embodiments are consistent and can be mutually referred. The technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship. The embodiments of the present application described below do not constitute a limitation on the protection scope of the present application.

[0067] To facilitate understanding of the technical solutions of the embodiments of the present application, first, a brief introduction of the related technologies of the present application is given as follows.

[0068] In a wireless communication network, as the diversification of service requirements and the enhancement of network functions, service implementation, network planning, configuration, and resource scheduling also become increasingly complex. For example, service implementation, network planning, configuration, and resource scheduling may involve modulation, coding, transmitters, receivers, multi-antenna technology, or positioning technology in a wireless communication system. Among them, the network device can implement schemes such as signal modulation and demodulation, information encoding and decoding, CSI feedback, BM or mobility management by performing related operations.

[0069] For example, the communication device can execute the technical solutions in the above-mentioned various service scenarios through a traditional algorithm, or can apply artificial intelligence (AI) / machine learning (ML) technology to the technical solutions in the above-mentioned service scenarios. AI / ML technology refers to model training through related data, so as to achieve a specific purpose using a trained model. The purpose that the model can achieve is related to the data used during training.

[0070] Among them, taking the service scenario of AI / ML-based CSI feedback as an example, the service scenario of AI / ML-based CSI feedback includes AI / ML-based CSI compression and AI / ML-based CSI prediction. AI / ML-based CSI compression refers to that the terminal compresses the downlink CSI (measured by the terminal) through AI / ML technology. Then, the terminal can send the compressed CSI to the network device through the air interface. The network device restores (decompresses) the CSI based on AI / ML. Compared with the traditional compression algorithm, the AI / ML-based compression algorithm has higher compression rate and better CSI restoration capability. Therefore, the terminal can feed back more CSI through smaller air interface overhead, so that the network device can more accurately perform downlink precoding. AI / ML-based CSI prediction refers to that the network device predicts the downlink CSI at a future time based on the downlink CSI at a current / historical time through AI / ML technology, and then performs precoding according to the predicted CSI. The CSI predicted by this scheme is more matched to the channel state when the downlink data is scheduled, so it can overcome the problem of channel aging and achieve more accurate downlink precoding. The training data of the model involved in the service scenario of AI / ML-based CSI feedback includes CSI.

[0071] Taking a business scenario of BM based on AI / ML as an example, a network device and a terminal device can predict a sending beam and / or a receiving beam through AI / ML technology, for example, infer a small amount of beam scanning results through AI / ML to obtain a preferred beam. Compared with a traditional scheme in which a large amount of beams need to be scanned to obtain a preferred beam, beam prediction based on AI / ML can reduce the processing overhead of beam scanning. For example, a terminal can scan a small amount of beams, and then predict a preferred beam from a large number of candidate beams through an AI / ML model, so that it is not necessary to scan all candidate beams, thereby reducing the overhead. The small amount of beams scanned by the terminal can be sparse beams or wide beams, and the candidate beams can be dense beams or narrow beams. The terminal can input beam information scanned at a current / historical moment into the model to predict a preferred beam at a future moment, so that it is not necessary to perform beam scanning again at the future moment, thereby improving the efficiency of beam scanning. Training data of a model involved in the business scenario of BM based on AI / ML includes beam information, for example, beam information can be beam ID and / or beam corresponding reference signal receiving power (RSRP) and other beam related information.

[0072] Taking a business scenario of positioning based on AI / ML as an example, a communication device inputs channel information into an AL / ML model to infer an intermediate parameter required for positioning, or directly obtains a position coordinate value. Compared with a traditional positioning algorithm, the intermediate parameter or the position coordinate value obtained based on AL / ML is more accurate. Training data of a model involved in the business scenario of positioning based on AL / ML includes channel information and / or position information, and the channel information includes power information, phase information, time delay information, distance information, speed information, channel scattering information, and line of sight (LOS) / non line of sight (NLOS) information and other channel related information.

[0073] In the process of implementing the above scheme through a wireless communication network, data transmission is usually involved between communication devices, for example, a network device needs to transmit required model data to a terminal, and the data transmission not only affects the actual execution effect of the terminal, but also generates communication overhead. However, in the actual process, the network side is difficult to evaluate the expected performance of the terminal side after the model delivery, which may lead to the model delivery strategy formulated by the network device being unable to guarantee high model expected benefits. For example, the network device is subject to communication resource constraints and can only deliver part of the model, but the delivered model does not achieve good model benefits, which leads to the model delivery strategy formulated by the network device being unable to achieve good model benefits, and at the same time, due to the occupation of a large amount of communication resources, the normal delivery of other models is affected. Based on this, the first node in the present application can send the second node related information representing the model delivery performance index of the first model, so that the second node can fully consider the model delivery performance of the first model (also referred to as the performance-price ratio of model delivery, or also referred to as the benefit value of model delivery, or also referred to as the target performance of model delivery, or also referred to as the stability of model delivery, or also referred to as the robustness of model delivery) when formulating the model delivery strategy. Therefore, the second node can reasonably plan the communication resources required for model delivery based on the related information of the model delivery performance index, avoid wasting communication resources, and at the same time achieve higher benefits of the model.

[0074] The technical scheme of the embodiments of the present application can be used in various communication systems, which can be a third generation partnership project (3GPP) communication system, for example, a long term evolution (LTE) system, a fourth generation (4G) system such as a new radio (NR) system, a 5G system, a system of mixed networking of LTE and 5G, a communication and sensing integrated system, a non-terrestrial network (NTN), a device-to-device (D2D) communication system, a vehicle to everything (V2X) communication system, a machine-type communication (MTC) system, an internet of things (IoT) system, or other future communication systems. The communication system can also be a non-3GPP communication system, which is not limited.

[0075] The communication system applicable to the present application is only illustrative, and the communication system applicable to the present application is not limited thereto. The communication system provided by the present application does not cause any limitation to the scheme of the present application. Herein, the following will not be described in detail.

[0076] FIG. 1 shows a possible, non-limiting system diagram. As shown in FIG. 1, a communication system 10 includes a radio access network (RAN) 100 and a core network (CN) 200. The RAN 100 includes at least one RAN node (e.g., 110a and 110b in FIG. 1, collectively referred to as 110) and at least one terminal (e.g., 120a-120j in FIG. 1, collectively referred to as 120). Other RAN nodes, such as wireless relay devices and / or wireless backhaul devices (not shown in FIG. 1), etc., can also be included in the RAN 100. The terminal 120 is connected to the RAN node 110 in a wireless manner. The RAN node 110 is connected to the core network 200 in a wireless or wired manner. The core network node in the core network 200 and the RAN node 110 in the RAN 100 can be different physical devices, or can be the same physical device integrated with the core network logic function and the radio access network logic function.

[0077] In a possible implementation, the core network node can refer to a device in the core network 200 that provides service support for the terminal 120. In the embodiments of the present application, the core network node in the core network 200 includes a sensing function (SF) network element, which is mainly used to implement sensing functions, such as sensing control functions and / or sensing calculation functions. Further, the SF network element can also support sensing billing functions when the terminal 120 and / or the RAN node 110 perform sensing. For example, the sensing control function can include determining sensing devices, sensing nodes, etc. The sensing device can be understood as a device that transmits and / or receives sensing signals, and further performs corresponding signal processing on the received echo signals to obtain sensing measurement data. For example, the sensing device can be the RAN node 110 or the terminal 120, etc. The sensing node can refer to a network node participating in the sensing service process in the wireless network. The sensing calculation function can include performing corresponding signal processing on the echo signals received by the sensing device to obtain sensing measurement data, and further processing the sensing measurement data and application information to obtain sensing results, etc.

[0078] For example, the SF network element can also be referred to as a communication device, for example, the SF network element can be understood as a communication device with a core network sensing function. In addition, the SF network element can also be referred to as a sensing server, etc., without limitation.

[0079] In a possible scenario, the functions of the SF network element can be implemented by a network data analytics function (NWDAF) network element, or the SF network element and the NWDAF network element can be combined.

[0080] Optionally, in addition to the SF network element, the core network nodes in the core network 200 can further include at least one of the following: an access and mobility management function (AMF) network element, a session management function (SMF) network element, a user plane function (UPF) network element, a policy control function (PCF) network element, a unified data management (UDM) network element, an application function (AF) network element, a network exposure function (NEF) network element, a network slice selection function (NSSF) network element, or a location management function (LMF) network element, and the like. Of course, the core network 200 can further include other core network nodes, which are not limited.

[0081] The AMF network element is a network element deployed in the core network 200, which provides mobility management and connection management for the network, such as user location update, user registration network, user handover, and the like. The AMF network element can serve as an intermediate route of the LMF, the SMF, and the RAN 100. The SMF network element is mainly responsible for session management in the mobile network, such as session establishment, modification, release, and the like. The UPF network element is a functional network element of the user plane, which is mainly responsible for connecting external networks and processing user packets, such as forwarding, charging, and the like. The PCF network element is mainly responsible for providing policies to the AMF and the SMF, such as quality of service (QoS) policies, slice selection policies, and the like. The UDM network element is used to store user data, such as subscription information, authentication / authorization information, and the like. The AF network element is responsible for providing services to the 3GPP network. The NEF network element is mainly used to open the capabilities of various network functions and is responsible for converting internal and external information. The LMF network element is a device or component deployed in the core network 200, which provides positioning functions for the terminal 120, for example, the LMF network element can initiate a positioning process and perform positioning on a specific terminal.

[0082] It should be noted that the network element in the present application can also be referred to as an entity or a functional entity, for example, the SF network element can also be referred to as an SF entity or an SF functional entity. In addition, the above-mentioned AMF network element, SMF network element, UPF network element, PCF network element, UDM network element, AF network element, NEF network element, LMF network element can also have other names in future communication systems, which are not limited in the present application.

[0083] In a possible implementation, the RAN 100 can be a 3rd generation partnership project (3GPP) related cellular system, for example, a 4G, 5G mobile communication system, or a future-oriented evolution system. The RAN 100 can also be an open RAN (O-RAN or ORAN), a cloud radio access network (CRAN), an NTN network (such as an NTN supporting a transparent mode and / or a regenerative mode, or an NTN supporting a gaze mode (earth fixed cell) and / or a non-gaze mode (earth moving cell)), or a wireless fidelity (WiFi) system. The RAN 100 can also be a communication system in which two or more of the above systems are fused.

[0084] The RAN node 110, which can also be referred to as an access network device, a RAN entity or an access node, etc., constitutes part of the communication system to help terminals to achieve wireless access. The plurality of RAN nodes 110 in the RAN 100 can be nodes of the same type or nodes of different types. In some scenarios, the roles of the RAN node 110 and the terminal 120 are relative, for example, the network element 120i in FIG. 1 can be a helicopter or a drone, which can be configured as a mobile base station. For those terminals 120j accessing the RAN 100 through the network element 120i, the network element 120i is a base station; but for the base station 110a, the network element 120i is a terminal. The RAN node 110 and the terminal 120 are sometimes referred to as communication apparatuses, for example, the network elements 110a and 110b in FIG. 1 can be understood as communication apparatuses with base station functions, and the network elements 120a-120j can be understood as communication apparatuses with terminal functions.

[0085] For the RAN node 110, in one possible scenario, the RAN node 110 can be a base station, an evolved Node B (eNodeB, also referred to as eNB), an access point (AP), a transmission reception point (TRP), a next generation NodeB (gNB), a next generation NodeB in a future mobile communications system, or an access node in a WiFi system, etc. The RAN node 110 can be a macro base station (e.g., 110a in Figure 1), a micro base station or indoor station (e.g., 110b in Figure 1), a relay node or donor node, or a wireless controller in a CRAN scenario. For example, a satellite base station, a radio network controller (RNC), a base station controller (BSC), a base transceiver station (BTS), a home base station (e.g., home eNodeB, or home NodeB, HNB), a relay station, a balloon station, a drone station, a wireless backhaul node, or a G node in a starlink, etc. It can be understood that the network device can be a device arranged on the ground, or a non-ground device (such as a satellite, a drone, a high-altitude communication device, etc.). In addition, in a communication system using different wireless access technologies, the name of the network device with base station function may be different, which is not limited in the present application. Optionally, the RAN node 110 can also be a server, a wearable device, a vehicle or a vehicle-mounted device, etc. For example, the access network device in the vehicle to everything (V2X) technology can be a road side unit (RSU). The RAN node 110 is also referred to as a next generation-RAN (NG-RAN) node.

[0086] In another possible scenario, a terminal is assisted by multiple RAN nodes 110 to implement wireless access in cooperation, and different RAN nodes 110 respectively implement part of the functions of a base station. For example, a RAN node 110 can be a central unit (CU, also known as a central unit), a distributed unit (DU, also known as a distributed unit), a CU-control plane (CP), a CU-user plane (UP), or a radio unit (RU), etc. The CU and the DU can be separately arranged, or can also be included in the same network element, such as a baseband unit (BBU). The RU can be included in a radio frequency device or a radio frequency unit, such as a remote radio unit (RRU), an active antenna processing unit (AAU), or a remote radio head (RRH).

[0087] In different systems, the CU (or CU-CP and CU-UP), DU or RU can also have different names, but those skilled in the art can understand their meanings. For example, in an ORAN system, the CU can also be referred to as an O-CU (open CU), the DU can also be referred to as an O-DU, the CU-CP can also be referred to as an O-CU-CP, the CU-UP can also be referred to as an O-CU-UP, and the RU can also be referred to as an O-RU. For the convenience of description, the CU, CU-CP, CU-UP, DU and RU are taken as examples for description in this application. Any one of the CU (or CU-CP, CU-UP), DU and RU in this application can be implemented by a software module, a hardware module, or a combination of a software module and a hardware module.

[0088] For the terminal 120, in a possible scenario, the terminal 120 can be a device for implementing a wireless communication function, for example, a terminal or a chip or circuit used in a terminal, or an entity associated with the terminal, etc. Among them, the terminal 120 can be a user equipment (UE), an access terminal, a terminal unit, a terminal station, a mobile station (MS), a mobile station, a remote station, a remote terminal, a mobile device, a wireless communication device, a terminal agent or a terminal device, a subscriber unit, a smart phone, a wireless data card, a tablet computer, a wireless modem, a laptop computer, a machine type communication (MTC) terminal, a tag, etc. in a 5G network or a future evolved public land mobile network (PLMN). The access terminal can be a cellular phone, a cordless phone, a session initiation protocol (SIP) phone, a wireless local loop (WLL) station, a personal digital assistant (PDA), a handset with wireless communication function, a computing device or other processing device connected to a wireless modem, a vehicle-mounted device or a wearable device, a virtual reality (VR) terminal, an augmented reality (AR) terminal, a wireless terminal in industrial control, a wireless terminal in self driving, a wireless terminal in remote medical, a wireless terminal in smart grid, a wireless terminal in transportation safety, a wireless terminal in smart city, a wireless terminal in smart home, or a terminal node (T node) in starlink, etc. In a possible implementation, the terminal 120 can be mobile or fixed. It can be understood that the terminal and the mobile user can be completely independent. All information related to the user can be stored in a subscriber identity module (SIM) card, which can be used on a terminal device.The terminal can send and / or receive signals through the air interface to complete interaction with the network side device.

[0089] The chip or circuit in the terminal includes at least one of components inside the terminal, such as a chip, a central processing unit (CPU), a network process unit (NPU), and a terminal radio frequency module.

[0090] The entity associated with the terminal includes a server on the terminal side, a computing / processing node, a computing / processing entity, a computing / processing unit, a server, such as an over the top (OTT) server, etc. OTT refers to a third party other than a network operator providing various services to users based on the operator's network, such as OTT voice communication services, OTT multimedia services, and OTT data processing services, etc. The terminal interacts with relevant information (such as data) through communication with the associated network entity. For example, the associated network entity and the terminal belong to the same manufacturer. Due to model training, model selection, etc., it can not be executed on the terminal, but on the OTT server on the terminal side, so the "terminal" in this embodiment also includes the OTT server on the terminal side.

[0091] It should be understood that the terminal in the embodiment can also be referred to as "terminal side" (UE side) or "terminal part" (UE part).

[0092] For example, as shown in FIG. 2, an exemplary implementation of the system shown in FIG. 1. The communication system can include an AI / ML node, a first terminal, and a first device. The first device can provide services for the terminal.

[0093] Optionally, the first device can be a server, which can be a single server, or a server cluster composed of multiple servers. In some embodiments, the server cluster can also be a distributed cluster. The server can provide services for the chip, so it can also be called a chip server. Alternatively, the first device can be a first network element in the core network. The first device can train a model or deliver a model or deliver a model inference result for the terminal it serves.

[0094] Optionally, the communication system shown in FIG. 2 can include a network device. The network device can be any device deployed in an access network that can communicate with a terminal (e.g., the first terminal, the second terminal) wirelessly, can also be a chip or chip system that can be disposed in the above-mentioned device, can also be a logical node or a logical module or a software-implemented function, and is mainly responsible for functions such as wireless physical control, resource scheduling, radio resource management, quality of service management, data compression and encryption, wireless access control, and mobility management. Specifically, the network device can be a device supporting wired access or a device supporting wireless access.

[0095] In some embodiments, when the model data is stored in the network device, the model can be directly transmitted to the first terminal by the network device. When the model data is stored in the first device, the model can be transmitted to the first terminal by the first device through UP interface data transmission, or the model can be transmitted to the network device by the server and then forwarded to the first terminal by the network device.

[0096] The AI / ML node in FIG. 2 is used to support the use of AI / ML technology in an AI / ML scenario.

[0097] Optionally, the AI / ML node can be deployed in one or more of the following positions in the communication system shown in FIG. 2: the network device, the first terminal, the second terminal, the first device, etc., or the AI / ML node can also be deployed separately, for example, in a position other than the above-mentioned devices.

[0098] For example, the AI / ML node can be deployed in a host or a cloud server of an OTT system. When the device deploying the AI / ML node communicates with the network device, the device can also act as a terminal in the communication system. When the device deploying the AI / ML node communicates with the terminal, the device can also act as a network device in the communication system.

[0099] It can be understood that the number of AI / ML nodes is not limited in the present application. For example, when there are multiple AI / ML nodes, the multiple AI / ML nodes can be divided based on functions, such as different AI / ML nodes being responsible for different functions.

[0100] It can also be understood that the AI / ML node can be a separate device, can also be integrated into the same device to implement different functions, or can be a network element in a hardware device, or can be a software function running on a dedicated hardware, or a virtualized function instantiated on a platform (e.g., a cloud platform), and the specific form of the AI / ML node is not limited in the present application.

[0101] The AI / ML node can be an AI / ML network element or an AI / ML module.

[0102] It can be understood that the above Figure 2 is only a schematic diagram, and does not constitute a limitation on the applicable scenarios of the technical solutions provided in the present application. It should be understood by those skilled in the art that, in the specific implementation process, the communication system shown in Figure 2 can also include fewer devices than those shown in Figure 2, or the communication system shown in Figure 2 can also include other devices, and the number of devices in the communication system shown in Figure 2 can also be determined according to specific needs, and is not limited.

[0103] Optionally, each device in Figure 2, such as the first device, the first terminal, the network device, and the second terminal, can also be referred to as a communication apparatus, which can be a general-purpose device or a special-purpose device, and the embodiments of the present application do not make specific limitations.

[0104] Optionally, the related functions of each device in Figure 2 of the present application can be implemented by one device, or can be implemented by multiple devices together, or can be implemented by one or more functional modules in a device, and the embodiments of the present application do not make specific limitations. It can be understood that the above functions can be network elements in a hardware device, or software functions running on a special-purpose hardware, or a combination of hardware and software, or virtualized functions instantiated on a platform (for example, a cloud platform).

[0105] In a possible implementation manner, the network device (such as an access node or a core network node) in the embodiments of the present application and the terminal 120 can also be referred to as a communication apparatus, which can be a general-purpose device or a special-purpose device, and the network device can include an access node (RAN node), an operation administration and maintenance (OAM) device, or a core network node. For the OAM device, it can include a device in an element management system (EMS), or a device in a network management system (NMS). It should be understood that the network device in the present embodiment can also be referred to as a "network side" or a "network part". The embodiments of the present application do not make specific limitations.

[0106] In a possible implementation, the related functions of the terminal 120 or the network device in the embodiments of the present application can be implemented by one device, or can be implemented by multiple devices together, or can be implemented by one or more functional modules in a device, and the embodiments of the present application do not make a specific limitation. It can be understood that the above functions can be network elements in a hardware device, can be software functions running on a dedicated hardware, or can be a combination of hardware and software, or can be a virtualized function instantiated on a platform (for example, a cloud platform).

[0107] It should be noted that the RAN node can be a device or a component in the device in the above NG-RAN, for example, can be an ng-eNB node, a gNB node, or a transmission point (TP) in the ng-eNB node and the gNB node, a transmission and reception point (TRP), or a central unit (CU) integrated on the NG-RAN. The RAN node can also be a network element with transmission function, for example, a transmission measurement function network element (TMF). In some embodiments, the RAN node can also be an access node in the O-RAN system. The RAN is usually composed of a series of modules, such as antenna, RRU, and BBU modules. The traditional RAN architecture defines the overall reception and output of the RAN node, and does not limit the transmission and contact between internal modules. The O-RAN architecture defines the architecture contact and standardized interface between each module in the RAN, so that the RAN can be decoupled into multiple standardized modules, thereby realizing the combination and replacement of the modules.

[0108] An exemplary structure of a possible, non-limiting O-RAN system is shown in FIG. 3. In this example, a service management and orchestration framework (SMO) is used as a network management device in the O-RAN to manage the operation of the devices in the O-RAN. A non-real time RAN intelligent controller (Non-RT RIC) is located in the SMO module to implement non-real time intelligent management of RAN functions, such as to implement AI / ML workflows including model training and model updating, and to guide applications / functions in the Near-RT RIC based on policies. A near-real time RAN intelligent controller (Near-RT RIC) is used to implement near-real time intelligent management of the RAN. Near-real time control and optimization of the modules and resources of the O-RAN are implemented through data collection and related operations on the E2 interface.

[0109] An O-RAN central unit (O-CU) includes an O-RAN central unit control plane (O-CU-CP) and an O-RAN central unit user plane (O-CU-UP). The O-CU is used to implement the radio resource control (RRC) layer, the packet data convergence protocol (PDCP) layer, and the service data adaptation protocol (SDAP) layer and other control functions. The O-CU-CP is used to implement the functions of the RRC layer and the control plane functions of the PDCP layer. The O-CU-UP is used to implement the functions of the SDAP layer and the user plane functions of the PDCP layer.

[0110] An O-RAN distributed unit (O-DU) is configured to implement a radio link control (RLC) layer, a media access control (MAC) layer, and a higher physical layer (Higher PHY). The Higher PHY functions include one or more of forward error correction (FEC) encoding / decoding, scrambling / descrambling, or modulation / demodulation.

[0111] An O-RAN radio unit (O-RU) is configured to implement lower physical layer (Lower PHY) functions and radio frequency functions. The Lower PHY functions include one or more of fast Fourier transform (FFT) transform / inverse fast Fourier transformation (iFFT) transform, digital beamforming, or extraction and filtering of a physical random access channel (PRACH). That is, the O-RU has functions similar to those of a radio frequency device such as a TRP or a RRH and Lower PHY processing functions. In addition, the O-RU, the O-CU, and the O-DU can be collectively referred to as an O-eNB / gNB and configured to implement the above functions.

[0112] An O-RAN cloud (O-Cloud) is a cloud computing platform that includes physical infrastructure nodes configured to host O-RAN functions such as RICs and O-DUs. The O-Cloud supports software components (such as operating systems, virtual machine monitors, and container runtimes), management, and orchestration functions.

[0113] In one possible scenario, the O-RAN system further includes a sensing unit (SU). The SU is configured to implement sensing-related functions, such as transmitting a sensing signal and / or receiving an echo signal of the sensing signal, performing signal processing on the received echo signal to obtain sensing measurement data, and performing sensing-related processing.

[0114] As one possible implementation, a RAN node can include at least one of a CU, a DU, a SU, and a RU. A communication interface exists between the CU and the SU. A communication interface can or can not exist between the SU and the DU. In the case where no communication interface exists between the SU and the DU, the SU and the DU can communicate through the CU.

[0115] In the O-RAN architecture, the module receiving the difference reporting between the twin channel and the measurement channel can be a CU, a RT RIC, a Non-RT RIC, etc., and the DU is responsible for receiving signals, signal processing, multipath measurement, and channel difference calculation.

[0116] For example, the O-RAN system includes communication interfaces between newly added internal components and other communication interfaces. For example, the A1 interface is an interface between the Non-RT RIC and the Near-RT RIC, which is used for intelligent and dynamic control of O-RAN internal wireless resources. The Non-RT RIC can provide policies, rich information, and ML model updates to the Near-RT RIC through the A1 interface, and the Near-RT RIC can provide policy feedback to the Non-RT RIC through the A1 interface.

[0117] The E2 interface is an open interface between two endpoints, which is used to connect the Near-RT RIC and the RAN node, including the CU, DU in 5G, the O-RAN compatible eNB in 4G, the O-CU (O-CU-CP and / or O-CU-UP) and / or O-DU in O-RAN, etc. The Near-RT RIC can obtain RAN node data collection and feedback through the E2 node, and the RAN node can obtain control feedback from the Near-RT RIC through the E2 node.

[0118] The O1 interface is an interface between the management entity in the SMO and the O-RAN module, which is used for operation management. Through this interface, network management (such as fault management, configuration management, billing management, performance management, security management, also known as FCAPS management), software management, and file management are realized. The O2 interface is an interface between the SMO and the infrastructure management framework supporting the O-RAN virtual network function.

[0119] The open front-haul (FH) CUS-Plane interface includes the control plane C-Plane, the user plane U-Plane, and the synchronization plane S-Plane interface. The control plane is used for real-time control between the O-DU and the O-RU, such as transmitting the weight for beamforming from the O-DU to the O-RU, or power control from the O-DU to the O-RU, etc. The user plane is used for transmitting communication data between the access network device and the terminal between the DU and the RU. The synchronization plane is used for the O-DU to provide clock synchronization to the O-RU. The Open FH M-Plane interface is a management plane interface, which is used for connection between the O-RU and the O-DU and the SMO, and can realize management, monitoring, and configuration functions, etc.

[0120] In addition, the NG interface is an interface between a RAN node (e.g., a base station, a CU, a CU-CP, a CU-UP) and a core network, NG-u is a user plane NG interface, and NG-c is a control plane NG interface. The Xn interface is an interface between NR RAN nodes, Xn-u is a user plane Xn interface, and Xn-c is a control plane Xn interface. The X2 interface is an interface between LTE RAN nodes, X2-u is a user plane X2 interface, and X2-c is a control plane X2 interface. In the NR system, the X2 interface is mainly used in an E-UTRA-NR dual connectivity scenario (EN-DC), in which a master base station is an LTE RAN node, and the master base station is connected to an LTE core network through the X2 interface. The E1 interface is an interface between a CU-CP and a CU-UP, the F1-c interface is an interface between a CU-CP and a DU, and the F1-u interface is an interface between a CU-UP and a DU.

[0121] In some embodiments, the communication apparatus in the embodiments of the present application can implement an AI / ML workflow (also referred to as model operation), which includes data collection, model training, model delivery, model updating, model inference, model monitoring, and model management.

[0122] Data collection refers to the collection of relevant data required by the communication apparatus before performing other operations of the model (model training, model updating, data analysis, model inference, and model monitoring). In the data collection process, the communication apparatus can divide data corresponding to different data features into different data sets and train different models, so that the model can better adapt to a certain specific feature in the inference stage and achieve better performance. Model training refers to training a trained model through collected training data. Model delivery, which can also be referred to as model transmission or model sending, refers to the delivery of the structure of the model and / or the parameters of the model. Model updating refers to updating a trained model through newly collected data, and the updating method is similar to retraining the model. Model inference refers to inputting collected data into a trained model and outputting corresponding result data. Model monitoring refers to monitoring the relevant state during model inference to detect the inference performance of the model. The inference performance of the model can be affected by environmental changes. When the channel environment changes, if the model can no longer well adapt to the channel environment, the reliability of the model will decrease. In order to ensure the reliability of the model, the network device and / or the terminal can use model monitoring to detect the performance of the model, and when the performance decreases, the network device and / or the terminal need to timely switch the model or deactivate the model. Model management refers to management operations in each stage of the model, such as model selection, model switching, model activation, and model deactivation.

[0123] It should be understood that the model involved in the embodiments of the present application can be described as a function (such as an AI function or an ML function), a feature or an algorithm, and the "model operation" can also be referred to as "function operation". The above model training, model delivery, model updating, model inference, model monitoring and model management can be replaced by function training, function updating, function inference, function monitoring (or performance monitoring) and function management respectively. The function here can be understood as a function corresponding to artificial intelligence. One model can implement one or more functions, and one or more models can also jointly implement one function.

[0124] The node performing the model inference can be a terminal, a network device, or jointly performed by the terminal and the network device. Therefore, the deployment mode of the model is divided into unilateral model deployment and bilateral model deployment.

[0125] The unilateral model deployment refers to that for an air interface feature, a use case or a functionality, only deploying an AI / ML model (referred to as a network device side AI / ML model) on the network device side for inference or deploying an AI / ML model (referred to as a terminal side AI / ML model) on the terminal side for inference can complete the entire inference process of the air interface feature. For example, the BM use case or the positioning technology use case can complete the entire inference process by deploying the AI / ML model unilaterally. The unilateral model includes a network side model and a UE side model. For the network side model, the terminal can report measurement information to the network device as input data of the network side model for model inference / training / monitoring / management; for the terminal side model, the terminal can perform model inference / training / monitoring based on the measurement information and report the inference result or the monitoring result to the network device. Whether it is the network side model or the terminal side model, for some use cases, the network device needs to indicate or configure corresponding resources (such as reference signal resources) to the terminal for measurement of the terminal to generate the above reported measurement information or for inference or monitoring or management of the terminal side model.

[0126] Bilateral model deployment refers to that for one air interface feature / use case / function, both the terminal and the network device deploy an AM / ML model, at this time the network device side model and the terminal side model need to be paired to complete the entire inference process of the air interface feature. Taking CSI compression as an example, the terminal side model infers the original CSI to achieve the effect of compression, and feeds back the compressed CSI data inferred to the network device through the air interface. After receiving the compressed CSI data, the network device infers the compressed CSI data through the network device side model to achieve the effect of decompression, and obtains the restored CSI. For the network device side model and the terminal side model with good matching degree, the restored CSI is closer to the original CSI. For the network device side model and the terminal side model with poor matching degree, the restored CSI is quite different from the original CSI.

[0127] For example, in combination with different deployment modes of the model and the relationship between the model and the function, the terminal and the network device can be applied to the following six scenarios when deploying the model or the function.

[0128] Scenario 1: The terminal side and the network side jointly deploy one model, and the model can implement at least one function. For example, the model can implement one or more of function 1, function 2 and function 3; wherein the terminal is used to execute function 1 in the model, and the network device is used to execute function 2 and / or function 3 in the model.

[0129] Scenario 2: The network side deploys one model, and the model can implement at least one function. For example, the model can implement function 1 and / or function 2; wherein the terminal can report measurement information required by the model to the network device, and the network device can execute function 1 and / or function 2 in the model.

[0130] Scenario 3: The terminal side deploys one model, and the model can implement at least one function. For example, the model can implement function 1 and / or function 2; wherein the terminal can obtain measurement information and execute function 1 and / or function 2 in the model.

[0131] Scenario 4: The terminal side and the network side jointly deploy one function, and the function is implemented through at least one model. For example, the function is implemented through one or more of model 1, model 2 and model 3; wherein the terminal is used to execute model 1, and the network device is used to execute model 2 and / or model 3.

[0132] Scenario 5: The network side deploys one function, and the function is implemented through at least one model. For example, the function is implemented through model 1 and / or model 2; wherein the terminal can report measurement information required by the model to the network device. The network device is used to execute model 1 and / or model 2.

[0133] Scenario 6, a terminal side deploys a function, which is implemented by at least one model. For example, the function is implemented by model 1 and / or model 2; wherein the terminal can obtain measurement information and invoke the execution of the model 1 and / or model 2 to implement the function.

[0134] In some embodiments, the model delivery can be triggered by the network device, or triggered by the terminal.

[0135] For example, as shown in FIG. 4, for the terminal side model in the unilateral model deployment, or the terminal side model in the bilateral model deployment, the model delivery method can be implemented by the following steps.

[0136] Step 401, the terminal sends model indication information to the network device. Correspondingly, the network device receives the model indication information from the terminal.

[0137] The model indication information is used to indicate the model that the terminal can support or can use or can apply. The model indication information can be supported model indication information, and can also be model availability indication information or model applicability indication information.

[0138] For example, the supported model indication information can be supported model indication information.

[0139] Step 402, the network device selects a model based on the model indication information.

[0140] For example, the network device can select the model that the terminal can support or can use or can apply from the model meta information.

[0141] Step 403, the network device sends model configuration information to the terminal. Correspondingly, the terminal receives the model configuration information from the network device.

[0142] The model configuration information includes the configuration parameters of the selected AI / ML model.

[0143] For example, the model configuration information can be RRC configuration information or RRC reconfiguration information. Then, the terminal and the network device can perform model transmission / delivery operation.

[0144] Step 404, the terminal sends model transmission / delivery waiting indication information to the network device. Correspondingly, the network device receives the model transmission / delivery waiting indication information from the terminal.

[0145] Exemplarily, the model transmission / delivery waiting indication information can be RRC configuration complete information or RRC reconfiguration complete information. The step 404 is an optional step, i.e., the step 404 can not be performed, and the terminal can not send the model transmission / delivery waiting indication information.

[0146] The step 405: the terminal and the network device perform a model transmission / delivery operation.

[0147] Exemplarily, if the model configured by the terminal is non-available, the required model can be obtained through the model transmission / delivery operation.

[0148] The step 406: the terminal sends model transmission / delivery completion indication information to the network device. Correspondingly, the network device receives the model transmission / delivery completion indication information from the terminal.

[0149] Exemplarily, the model transmission / delivery completion indication information can be RRC configuration complete information or RRC reconfiguration complete information. The step 406 is an optional step, i.e., the step 406 can not be performed, and the terminal can not send the model transmission / delivery completion indication information.

[0150] In some embodiments, the model delivery process involved in the present application is applicable to various communication systems. Taking the O-RAN system as an example, as shown in FIG. 5, in the case of a network device including a CU-CP, a CU-machine learning plane (MLP), and a DU, the CU-CP sends the model transmission / delivery operation to the terminal, and the terminal receives the model transmission / delivery operation from the CU-CP.

[0151] The step 501: the terminal sends terminal capability information to the CU-CP. Correspondingly, the CU-CP receives the terminal capability information from the terminal.

[0152] As an example, the terminal capability information is UE capability information.

[0153] The step 502: the CU-CP determines whether to use an ML function.

[0154] As an example, the ML function can also be referred to as ML functionality.

[0155] Step 503, in case the CU-CP determines to use ML function, the CU-CP sends a terminal context setup request message to the CU-MLP. Correspondingly, the CU-MLP receives the terminal context setup request message from the CU-CP.

[0156] As an example, the terminal context setup request message is: UE context setup request.

[0157] Step 504, the CU-MLP sends a model setup request message to the DU. Correspondingly, the DU receives the model setup request message from the CU-MLP.

[0158] As an example, the model setup request message is: ML model setup request.

[0159] Step 505, the DU sends a model setup response message to the CU-MLP. Correspondingly, the CU-MLP receives the model setup response message from the DU.

[0160] As an example, the model setup response message is: UE context setup response.

[0161] Step 506, the CU-MLP sends a terminal context setup response message to the CU-CP. Correspondingly, the CU-CP receives the terminal context setup response message from the CU-MLP.

[0162] As an example, the terminal context setup response message is: UE context setup response.

[0163] Step 507, the CU-CP sends an RRC reconfiguration request message to the terminal. Correspondingly, the terminal receives the RRC reconfiguration request message from the CU-CP.

[0164] Optionally, the RRC reconfiguration request includes an ML configuration message.

[0165] As an example, the RRC reconfiguration request message is: RRC reconfiguration request.

[0166] The ML configuration message is: ML configuration information.

[0167] Step 508, the terminal sends an RRC reconfiguration complete message to the CU-CP. Correspondingly, the CU-CP receives the reconfiguration complete message from the terminal.

[0168] As an example, the RRC reconfiguration complete message is: RRC reconfiguration complete.

[0169] Step 509, the terminal downloads the ML model. In some embodiments, in the case that the ML model data is stored in the access node, the access node can deliver the model data to the terminal, and in the case that the ML model data is stored in the cloud server, the cloud server can deliver the model data to the terminal through the access node.

[0170] Step 510, the terminal sends an ML model ready for use message to the CU-CP. Correspondingly, the CU-CP receives the ML model ready for use message from the terminal.

[0171] As an example, the ML model ready for use message is: ML model ready for use.

[0172] Step 511, the CU-CP sends the ML model ready for use message to the CU-MLP. Correspondingly, the CU-MLP receives the ML model ready for use message from the CU-CP.

[0173] Step 512, the CU-CP sends an ML model activation message to the terminal. Correspondingly, the terminal receives the ML model activation message from the CU-CP.

[0174] As an example, the ML model activation message is: Activate ML model.

[0175] It should be noted that the system described in the embodiments of the present application is to make the technical solution of the embodiments of the present application more clear, and does not constitute a limitation on the technical solution provided by the embodiments of the present application. Those skilled in the art can know that with the evolution of network architecture and the appearance of new business scenarios, the technical solution provided by the embodiments of the present application is also applicable to similar technical problems.

[0176] The communication method provided by the embodiments of the present application will be described below in combination with the communication system shown in FIG. 1, taking the interaction between the first node and the second node as an example. It should be noted that in the following embodiments of the present application, the message name between the first node and the second node, the name of each parameter, or the name of each information, etc. is only an example, and in other embodiments, it can also be other names, and the method provided by the present application does not make specific limitations.

[0177] It can be understood that, in the embodiments of the present application, each communication device (including the first node or the second node) can perform part or all of the steps in the embodiments of the present application, and these steps or operations are only examples, and the embodiments of the present application can also perform other operations or variations of various operations. In addition, each step can be performed in a different order as presented in the embodiments of the present application, and it is possible that not all operations in the embodiments of the present application are performed.

[0178] It can be understood that, in the embodiments of the present application, the first node and the second node are taken as an example to illustrate the execution subject of the interaction, but the present application does not limit the execution subject of the interaction. For example, the method executed by the first node in the present application can also be executed by a module (such as a chip, a chip system, or a processor) applied to the first node, and can also be realized by a logical node, a logical module or software that can realize all or part of the function of the first node. Or for example, the method executed by the second node in the present application can also be executed by a module (such as a chip, a chip system, or a processor) applied to the second node, and can also be realized by a logical node, a logical module or software that can realize all or part of the function of the second node.

[0179] The communication method provided by the embodiments of the present application is described below. As shown in FIG. 6, the communication method can include the following steps:

[0180] Step 601, the first node obtains first information.

[0181] For example, the first node can be a terminal in the above-mentioned communication system or a chip or circuit used in the terminal, or an entity associated with the terminal, etc. Alternatively, the first node can also be a network device in the above-mentioned communication system, such as an access node or a core network node, and the second node can also be a chip or circuit in the network device, or an entity associated with the network device, etc.

[0182] For example, the first information can be generated by the first node according to the data monitored by itself, or can be obtained by the first node from a third party. The first model can be a new model generated through model training, or the first model can also be an enhancement of the current model, and the embodiments of the present application do not make specific limitations.

[0183] The first model is used to enable AI / ML function, and the model delivery performance index information is used to represent the model delivery performance of the first model. The model delivery performance of the first model is also called the model delivery performance ratio of the first model, which can be used to evaluate the relationship between the communication resources occupied by model delivery and the expected benefits of the model.

[0184] In some embodiments, the model transmission performance indicator information can also be determined separately for each cell, that is, the corresponding model transmission performance indicator information is calculated and counted separately when the same first node is in different cells.

[0185] In the embodiments of the application, when the model transmission performance of the model deployed at the terminal side in unilateral deployment is poor, the second node needs to consume a large overhead when transmitting the model, and it is also difficult to ensure that the model can generate high benefits. When the model transmission performance of the model deployed in bilateral deployment is poor, due to the inability to guarantee the expected benefits of the model, in addition to the above problems, the network device will also waste AI computing power and power consumption and other resources.

[0186] In the embodiments of the application, model transmission will generate overhead, and frequent transmission of models with poor model transmission performance will cause waste of air interface resources and increase the complexity of the terminal in receiving and storing the model. The model transmission performance of the first model is affected by two aspects, the communication resources occupied by the model transmission and the expected benefits of the model. The communication resources occupied by the model transmission are determined by the data size of the model itself and the number of model transmissions in actual use. The expected benefits of the model are determined by the use effect of the model and the use duration of the model.

[0187] Among them, when the storage condition of the first node is good (for example, the available storage space is large), the first node can store the transmitted model data for a long time to realize the reuse of the model data, thereby reducing the number of model transmissions. Different models are suitable for different scenarios, and the adaptability of the scenario to the model determines whether the model is effective and the actual effect of the model. Therefore, the higher the recognition accuracy of the scenario, the more suitable the selected first model is to the actual scenario, and the better the actual use effect of the corresponding first model. In addition, the generalization of the first model determines the duration of the adaptation of the first model, and the better the generalization, the longer the first model can be used continuously in actual use. In summary, the model transmission performance of the first model is related to the storage condition of the first node, the recognition accuracy of the scenario, and the generalization of the first model.

[0188] In some embodiments, the model transmission performance indicator information of the first model can be determined by an entity that determines the model performance, for example, when the first model is deployed at the terminal, the model performance of the first model is determined by the terminal or by the network device. At this time, the model transmission performance indicator information of the first model is determined by the terminal.

[0189] When the first model is deployed at the network device, the model performance of the first model is determined by the network device or by the terminal. At this time, the model transmission performance indicator information of the first model is determined by the network device.

[0190] When the first model is deployed in the terminal and the network device at the same time, the model performance of the first model can be determined by the terminal, the network device, or both the terminal and the network device. At this time, the model transfer performance index information of the first model can be determined by the terminal, the network device, or both the terminal and the network device.

[0191] In addition, the statistical entity of the model transfer performance index information can be the terminal or the network device. For example, the terminal statistics can be for various models of specific terminal manufacturers, making the statistical results more robust; the network device can combine the model transfer performance index information from terminals of different manufacturers, which is suitable for more general scenarios. When the model performance determination entity and the statistical entity of the model transfer index information are different entities, the model performance determination entity needs to transfer the model performance to the statistical entity of the model transfer index information for generating the model transfer index information.

[0192] It should be understood that the above description of the model can also be applicable to functions, characteristics, or algorithms, etc. The model can use different expressions in different scenarios.

[0193] Step 602, the first node sends the first information to the second node. Correspondingly, the second node receives the first information from the first node.

[0194] For example, the second node can be a network device in the above communication system, such as an access node or a core network node, or a chip or circuit in the network device, or an entity associated with the network device, etc. Alternatively, the first node can also be a terminal in the above communication system or a chip or circuit used in the terminal, or an entity associated with the terminal, etc.

[0195] In some embodiments, the first node can report the first information using the capability information, or report the first information when the first node model is registered in the second node, or report the first information when the first node reports the supported model. The first information can be carried in a dedicated information, or can be carried in an RRC message, such as an RRC message carrying the capability information of the terminal device or the terminal assistance information (UE assistance information, UAI)

[0196] Wherein, the first information related introduction can refer to the above step 601, which will not be repeated here.

[0197] Step 603, the second node determines the model transfer strategy according to the first information.

[0198] In some embodiments, the second node can determine the management policy according to the first information and current network conditions. For example, the network conditions include information such as channel environment, communication quality, or running load. For example, the second node can determine whether and how to trigger the model delivery (for example, determine the quantization manner of the delivered model) in combination with the expected or historical yield of the model and the overhead of the model delivery. For example, for a model with a high expected use frequency after the model delivery, the priority of the corresponding model delivery is high. For a model with a low expected use frequency after the model delivery, the priority of the corresponding model delivery is low.

[0199] For example, when the second node is a network device in the communication system, the model delivery strategy can be that the second node directly sends the model data, or the model delivery strategy can be that the second node requests the opposite end to send the model data and waits for the response of the opposite end. When the second node is a terminal in the communication system, the model delivery strategy can be that the second node requests the terminal to send the model data.

[0200] Currently, it is difficult for the network side to evaluate the expected performance of the terminal side after the model delivery, which can cause the model delivery strategy formulated by the network device to fail to guarantee a high model expected yield. For example, the network device is limited by communication resources and can only deliver part of the models, but the delivered models do not achieve a good model yield, which causes the model delivery strategy formulated by the network device to fail to achieve a good model yield, and meanwhile, a large amount of communication resources are occupied, which affects the normal delivery of other models. Based on this, the first node can send the second node information representing the model delivery performance indicator of the first model, so that the second node can fully consider the model delivery performance of the first model (also referred to as the performance-cost ratio of the model delivery, or also referred to as the yield value of the model delivery, or also referred to as the target performance of the model delivery, or also referred to as the stability of the model delivery, or also referred to as the robustness of the model delivery) when formulating the model delivery strategy. Therefore, the second node can reasonably plan the communication resources required for the model delivery based on the information of the model delivery performance indicator, avoid wasting communication resources, and at the same time achieve a high yield of the model.

[0201] As a possible embodiment, the model delivery performance indicator information can be represented in various ways.

[0202] In some embodiments, the model delivery performance indicator information includes at least one model delivery performance indicator, and the at least one model delivery performance indicator includes at least one model delivery performance indicator value and / or at least one first value range.

[0203] In the embodiments of the present application, the model delivery performance indicator value refers to the specific numerical value corresponding to the model delivery performance indicator. For example, the model delivery performance of the first model is greater than 0.9.

[0204] The model delivery performance indicator value is generally used to represent the bottom line value reached by the model delivery performance indicator. Similarly, the first value range refers to the value interval corresponding to the model delivery performance indicator. The model delivery performance indicator and the model delivery performance can be positively correlated or negatively correlated. For example, the larger the value of the model delivery performance indicator, the better the model delivery performance. Alternatively, the larger the value of the model delivery performance indicator, the worse the model delivery performance. Since the model delivery performance of the model is affected by channel environment changes, communication configuration adjustments, or external physical interference, in this case, the first node can report the model delivery performance related information of the model in the form of a value or a value range according to the current situation, so that the second node can consider the influence of various factors when determining the model delivery strategy, and ensure the normal use of the model.

[0205] For example, the at least one model delivery performance indicator value includes at least two model delivery performance indicator values. The at least one first value range includes at least two first value ranges. In this way, the first information can include multiple values / value ranges of the model delivery performance indicator, so that the second node has more selection space when determining the model delivery strategy.

[0206] In some embodiments, the model delivery performance indicator information includes a value of the model delivery performance indicator and an indicator type.

[0207] The value can be a specific numerical value or a value range. The indicator type can be represented by a field, or an identifier, or a mapping table. For example, the first node can directly report the field information of the indicator type, or report an identifier that has a mapping relationship with the indicator type. The second node can determine the indicator type indicated by the identifier through the mapping table.

[0208] For example, when the first node needs to send multiple model delivery performance indicators of the first model to the second node, the model delivery performance indicator information can include values of multiple model delivery performance indicators and indicator types corresponding to the multiple model delivery performance indicators, so as to facilitate the second node to determine each model delivery performance indicator and the corresponding value.

[0209] In some embodiments, the model delivery performance indicator information includes a model satisfying the model delivery performance indicator or an identifier of the model.

[0210] For example, the first node can report the identifier of the model satisfying the model delivery performance indicator according to the indication of the second node, and the second node can directly determine the corresponding model delivery strategy based on the reported identifier of the model.

[0211] As a possible embodiment, the model delivery performance indicator information includes a mathematical statistic used to represent the model delivery performance of the first model.

[0212] In some embodiments, the model transfer performance indicator information comprises a mathematical statistic of the model transfer performance indicator for characterizing the first model.

[0213] The mathematical statistic can also be referred to as a probability statistic, such as a ratio, a mean / variance / standard deviation, or a cumulative distribution function (CDF) / probability mass function (PMF) / probability density function (PDF) of the model transfer performance, etc.

[0214] For example, the mathematical statistic of the model transfer performance can be:

[0215] 1. The probability of the model transfer performance < 0.7 is 10%;

[0216] 2. The probability of the model transfer performance between 0.7 and 0.8 is 20%;

[0217] 3. The probability of the model transfer performance between 0.8 and 0.9 is 60%;

[0218] 4. The probability of the model transfer performance > 0.9 is 10%.

[0219] That is, the model transfer performance can satisfy a specific probability distribution, and different intervals of the model transfer performance correspond to corresponding mathematical statistics. In addition, in addition to the probability representation, the mathematical statistic can also be represented by the mean / variance / standard deviation, or x% CDF / PMF / PDF, where x is a non-negative number, which will not be repeated. In this way, the present application can determine the corresponding model transfer performance indicator information by statistically analyzing the model transfer performance indicator, so that the model transfer performance indicator information can more accurately reflect the model transfer performance of the first model.

[0220] In some embodiments, the first information comprises at least one model transfer performance indicator information, and one or more model transfer performance indicator information in the at least one model transfer performance indicator information has a corresponding data feature or data feature identifier.

[0221] In some embodiments, the at least one model transfer performance indicator information one-to-one corresponds to the at least one data feature or data feature identifier.

[0222] In some embodiments, the at least one model transfer performance indicator one-to-one corresponds to the at least one data feature or data feature identifier.

[0223] Exemplarily, the first model corresponds to at least one model transfer performance indicator information, and the at least one model transfer performance indicator information includes first model transfer performance indicator information, and the first model transfer performance indicator information corresponds to a first data feature or a first data feature identifier.

[0224] Exemplarily, the at least one model transfer performance indicator information further includes second model transfer performance indicator information, and the second model transfer performance indicator information corresponds to a second data feature or a second data feature identifier.

[0225] For example, the at least one model transfer performance indicator information includes model transfer performance indicator information 1 and model transfer performance indicator information 2, and the at least one data feature includes data feature A and data feature B, where the model transfer performance indicator information 1 corresponds to the data feature A, and the model transfer performance indicator information 2 corresponds to the data feature B.

[0226] For another example, the at least one model transfer performance indicator information includes model transfer performance indicator information 1, model transfer performance indicator information 2, and model transfer performance indicator information 3, and the at least one data feature identifier includes data feature identifier A, data feature identifier B, and data feature identifier C, where the model transfer performance indicator information 1 corresponds to the data feature identifier A, the model transfer performance indicator information 2 corresponds to the data feature C, and the model transfer performance indicator information 3 corresponds to the data feature B.

[0227] That is, the data features or data feature identifiers corresponding to different model transfer performance indicator information can be the same or different. Different model transfer performance indicators in the at least one model transfer performance indicator correspond to different data features or data feature identifiers. Alternatively, part of the model transfer performance indicator information in the at least one model transfer performance indicator information does not have corresponding data features or data feature identifiers.

[0228] The "feature" in the embodiments of the present application can be replaced by: condition, situation, environment or circumstance, type, status, data, or data distribution. The "feature" in other terms "xx feature" (for example, "reference feature", "physical feature", or "feature of the x data") in the embodiments of the present application can be similarly replaced. Alternatively, the "feature" can also be referred to as "data feature" or "data classification feature".

[0229] Optionally, the data feature identifier can be referred to as an associated ID, a data ID, a dataset ID, a data categorization ID, or a property ID. The data feature identifier can be used by the terminal to classify measurement quantities used to determine model inputs or training data (e.g., model outputs, labels) used for model training, and can also be used for inference stage related operations. For example, the terminal can assume that a set / list of downlink transmission beams has the same or similar data features under the same data feature identifier.

[0230] In the embodiments of the present application, the feature can be used to represent the characteristics of the signals, channels, or data (e.g., data used for training the first model or data used for inference / processing of the first model) collected by the target node (e.g., the first node or the second node) during communication or during execution of the first model due to one or more of the device configuration or environmental factors such as node deployment, antenna form, and transceiver waveform. Different nodes may

[0231] In some embodiments, the first node can also send the data feature or the data feature identifier to the second node. The first node can add the data feature or the data feature identifier to the first information and send it to the second node through the first information. For example, the first information also includes the data feature or the data feature identifier. The first node can also send the data feature or the data feature identifier to the second node separately.

[0232] In some embodiments, the data feature is used to represent the characteristics of the first data corresponding to the first model.

[0233] In some embodiments, the data feature is used to represent the characteristics of the first data corresponding to the first model.

[0234] The data in the embodiments of the present application includes data obtained in a communication network, such as signal processing information, channel information, and radio frequency information, etc. The signal processing information includes information generated in a baseband signal processing process, such as information generated in a signal sampling, modulation, demodulation, coding, decoding, precoding, resource mapping, and / or digital filtering process. The channel information includes information corresponding to a channel environment, such as at least one or a combination of at least two of power information, amplitude information, phase information, time delay information, multipath information, signal propagation time information, distance information, speed information, large-scale channel information, small-scale channel information, channel scattering information, LOS / NLOS information, etc. For example, the data in the CSI use case and the BM use case is embodied as channel information CSI, and the data in the positioning use case is embodied as channel information and / or position information. The radio frequency information includes information generated in an analog processing process, such as information generated in a digital-to-analog conversion, analog-to-digital conversion, digital pre-distortion, frequency conversion, radio frequency modulation, radio frequency demodulation, power amplification, low-noise amplification, analog filtering, and / or duplex processing.

[0235] Taking the CSI feedback or beam management as an example, the first data includes CSI information. Optionally, the CSI information includes at least one of the following: channel response information, channel quality information, and beam information.

[0236] The channel response information can be at least one of a precoding matrix indicator (PMI), a precoding matrix, a precoding vector, an eigenvector, an eigenmatrix, a channel vector, a channel matrix, layer information (LI), and rank information (RI). The channel quality information can be at least one of an RSRP, an SINR, and a channel quality indicator (CQI). The beam information can be at least one of a CRI and a synchronization signal block resource indicator (SSBRI). The beam information can also be referred to as reference signal identification information.

[0237] As a possible embodiment, the model transmission performance indicator information is determined by state information and / or model transmission information. The state information includes first valid state information and / or first invalid state information.

[0238] In some embodiments, the model transmission information includes first model transmission quantization information, and the first model transmission quantization information includes a model transmission number and / or a model transmission time.

[0239] The model transmission times refer to the transmission times of the first model, and the model transmission time refers to the length of time occupied by the transmission of the first model. When the first model is transmitted multiple times, the model transmission time can be the sum of the length of time occupied by each transmission. In this way, the state information can represent the model expected benefit of the first model, and the model transmission information can represent the overhead of the model transmission of the first model. Therefore, the model transmission performance indicator information can be determined by the state information and / or the model transmission information.

[0240] For example, when the model transmission performance indicator information is determined by the state information, the greater the model expected benefit of the first model represented by the state information, the better the model transmission performance of the first model; on the contrary, the smaller the model expected benefit of the first model represented by the state information, the worse the model transmission performance of the first model. When the model transmission performance indicator information is determined by the model transmission information, the smaller the overhead of the model transmission of the first model represented by the model transmission information (for example, the fewer the model transmission times), the better the model transmission performance of the first model; on the contrary, the greater the overhead of the model transmission of the first model represented by the model transmission information, the worse the model transmission performance of the first model. When the model transmission performance indicator information is determined by the state information and the model transmission information, the model expected benefit of the first model and the overhead of the model transmission of the first model can be comprehensively evaluated, so as to more accurately reflect the model transmission performance of the first model.

[0241] In some embodiments, the first effective state information corresponds to an effective state of the first model, and the effective state includes at least one of the following:

[0242] a state in which the performance of the first model reaches or exceeds a first target performance indicator;

[0243] an active state of the first model; or

[0244] a supportable / applicable / available state of the first model;

[0245] And / or, the first invalid state information corresponds to an invalid state of the first model, and the invalid state includes at least one of the following:

[0246] a state in which the performance of the first model does not reach or exceed a second target performance indicator;

[0247] an inactive state of the first model; or

[0248] a non-supportable / non-applicable / non-available state of the first model.

[0249] That is, the first effective state information is used to represent information of the first model in the effective state, and the first effective state information is used to represent information of the first model in the invalid state.

[0250] Optionally, in embodiments of the present application, the target performance can correspond to inference performance, prediction performance, or monitoring performance of the first model. The inference performance or the prediction performance refers to the running performance of the first model when it is invoked (or activated / configured / enabled / triggered / executed / run), and the monitoring performance refers to the monitoring performance of the first model when it is invoked.

[0251] In some embodiments, when the target performance corresponds to the inference performance or the prediction performance of the first model, the target performance can be the system performance of the terminal when the first model is invoked, such as throughput, RSRP, signal noise ratio (SNR) / signal to interference plus noise ratio (SINR), or block error rate (BLER), and the like, which are parameters for characterizing the system performance.

[0252] Optionally, in embodiments of the present application, the target performance can also be the performance of the first model when it is running, such as running accuracy. For example, for a model used for CSI compression feedback, the target performance can be the restoration accuracy of the compressed CSI. For a model used for CSI prediction, the target performance can be the accuracy of CSI prediction. For a model used for beam prediction, the target performance can be the accuracy of beam prediction. For a model used for positioning inference, the target performance can be the accuracy of positioning inference. In addition, the first model can also be one of modulation, demodulation, encoding, decoding, channel equalization, channel estimation, pilot generation, precoding, resource mapping, resource demapping, interference suppression, interference estimation, interference prediction, receiver, or transmitter, or a combined model of multiple models, and the like, which will not be listed one by one.

[0253] It should be understood that the accuracy herein is also referred to as an intermediate key performance indicator (KPI), or prediction accuracy information. The accuracy is used to represent the similarity between the model inference result and the corresponding true value, such as generalized cosine similarity (GCS), squared generalized cosine similarity (SGCS), normalised mean square error (NMSE), beam information prediction accuracy, reference signal identification prediction accuracy, RSRP prediction accuracy, or positioning information prediction accuracy, etc. Among them, the true value corresponds to the measured measurement result. For example, the CSI restoration accuracy of the CSI compression feedback or the accuracy of the CSI prediction described above can be represented by SGCS or NMSE, the accuracy of the beam prediction can be represented by the comparison information (such as the confusion matrix) of the predicted result and the true value or label of the RSRP / channel state information-reference signal (CSI-RS) resource indicator (CRI). The accuracy of the positioning inference can be represented by the comparison of the predicted result and the real position coordinates or positioning label.

[0254] Optionally, in the embodiments of the present application, the target performance can also be a relative performance compared with the performance of the baseline model. For example, the performance of an algorithm for a certain model can be taken as the baseline performance in the present application, which can be represented in the above manner, and the target performance can be a relative performance compared with the baseline performance.

[0255] In some embodiments, when the target performance corresponds to the monitoring performance of the first model, the target performance can be the monitoring accuracy of the first model. For example, for the model for CSI compression feedback, the target performance can be the monitoring accuracy when monitoring the performance of the CSI compression feedback model. For the model for CSI prediction, the target performance can be the monitoring accuracy when monitoring the performance of the CSI prediction model, for example, the monitoring accuracy can be obtained by comparing the SGCS value corresponding to the performance monitoring result of the CSI prediction model with the SGCS value corresponding to the true performance result of the model (for example, obtaining the difference value of the two SGCS).

[0256] Optionally, in the embodiments of the present application, the first target performance indicator and the second target performance indicator can be the same or different. The performance of the first model reaching or exceeding the first target performance indicator includes that the performance of the first model reaching or exceeding the value of the first target performance indicator or being within the value range of the first target performance indicator. The performance of the first model not reaching or not exceeding the second target performance indicator includes that the performance of the first model not reaching or not exceeding the value of the second target performance indicator or not being within the value range of the second target performance indicator.

[0257] Since the performance of the first model will change due to various factors in actual use, in the embodiments of the present application, the value or the value range can be divided into multiple intervals, for example, the performance interval of the performance of the first model reaching or exceeding the value of the first target performance indicator is an interval corresponding to an effective state, or the performance interval of the value range of the first target performance indicator is an interval corresponding to an effective state; the performance interval of the performance of the first model not reaching or not exceeding the value of the second target performance indicator is an interval corresponding to an ineffective state, or the performance interval of the value range of the second target performance indicator is an interval corresponding to an ineffective state. In this way, the current state of the first model can be determined based on the interval in which the performance of the first model is located.

[0258] Optionally, in the embodiments of the present application, the activated state means that the model can be referred to as a configuration model, an enabled model, and the model starts to execute / run, etc. The model in the activated state means that the device can obtain the corresponding output result by running the model within a specific time according to the model input.

[0259] Optionally, in the embodiments of the present application, the supported state means that the device has the ability to run a specific model. For example, the terminal reports a supported model, which means that the terminal can run the model in the case of obtaining the model, that is, the terminal can obtain the corresponding output result by running the model within a specific time according to the model input.

[0260] Optionally, in the embodiments of the present application, the available state means that the device can obtain and run the model. For example, the terminal locally stores model-A and can run model-A, and for another example, the terminal can obtain model-A from a network device or a cloud server and can run model-A.

[0261] Optionally, in embodiments of the present application, the applicable state means that the model is ready for activation / inference / prediction. For example, if model-A is applicable to scenario-X, then when the current scenario is scenario-X, model-A is in the applicable state. The non-activation state and the non-supportable / non-applicable / non-available state are opposite states of the above-mentioned activation state and supportable / applicable / available state, and will not be described again.

[0262] In some embodiments, the first effective state information includes first effective quantification information of the first model, the first effective quantification information including effective time information and / or effective frequency information, wherein the effective time information includes a length of time during which the first model is in the effective state, and the effective frequency information includes a number of times during which the first model is in the effective state.

[0263] And / or, the first invalid state information includes first invalid quantification information of the first model, the first invalid quantification information including invalid time information and / or invalid frequency information, wherein the invalid time information includes a length of time during which the first model is in the invalid state, and the invalid frequency information includes a number of times during which the first model is in the invalid state.

[0264] In this way, the present application can quantize the effective state information and the invalid state information according to different dimensions, such as the time dimension and the statistical frequency dimension, so as to evaluate the model transmission performance index information of the first model.

[0265] Optionally, in embodiments of the present application, the length of time during which the first model is in the effective state and the length of time during which the first model is in the invalid state can be determined according to one or more monitoring resources configured. For example, the first node can monitor the one or more monitoring resources configured in the time period T1, determine whether the first model is in the effective state, if the first model is in the effective state, record that the length of time during which the first model is in the effective state increases by T1, and correspondingly, if the first model is in the invalid state, record that the length of time during which the first model is in the invalid state increases by T1.

[0266] Optionally, in embodiments of the present application, the number of times during which the first model is in the effective state and the number of times during which the first model is in the invalid state can be determined according to one or more monitoring resources configured. For example, the first node can monitor the one or more monitoring resources configured, determine whether the first model is in the effective state, if the first model is in the effective state, record that the number of times during which the first model is in the effective state increases by one, and correspondingly, if the first model is in the invalid state, record that the number of times during which the first model is in the invalid state increases by one.

[0267] In some embodiments, the model delivery performance indicator information comprises at least one of: first ratio information, first difference information, and first probability statistical information. The model delivery performance indicator information can be represented by the ratio information, the difference information, and the probability statistical information.

[0268] For example, the first ratio information can be represented by a ratio of valid time information and model delivery time, a ratio of valid time information and model delivery times, a ratio of valid times information and model delivery time, a ratio of valid times information and model delivery times, a ratio of invalid time information and model delivery time, a ratio of invalid time information and model delivery times, a ratio of invalid times information and model delivery time, or a ratio of invalid times information and model delivery times.

[0269] That is, the model delivery performance indicator information in the embodiments of the present application can be calculated based on various combinations between the first valid quantization information and / or the first invalid quantization information and the model delivery information. The model delivery performance indicator information can be positively or negatively correlated with the model delivery performance of the first model. For example, the larger the model delivery performance indicator information, the better the model delivery performance of the first model; the smaller the model delivery performance indicator information, the worse the model delivery performance of the first model. Alternatively, the larger the model delivery performance indicator information, the worse the model delivery performance of the first model; the smaller the model delivery performance indicator information, the better the model delivery performance of the first model.

[0270] In some embodiments, when the target performance indicator is default, the valid time information can be a length of time that the model is in an active state, a length of time that the model is in a supportable state, a length of time that the model is in an applicable state, or a length of time that the model is in a usable state. When the target performance indicator exists, the valid time information can be a length of time that the performance of the first model reaches or exceeds the first target performance indicator.

[0271] For example, the valid time information can be one or more of a length of time that the model is in an active state, a length of time that the model is in a supportable state, a length of time that the model is in an applicable state, or a length of time that the model is in a usable state. The valid time information can be represented as T0. The model delivery times can be represented as N0. The length of time can be a continuous period of time, or a length of time composed of multiple continuous periods of time. The model delivery performance indicator information At this time, the model delivery performance indicator information is negatively correlated with the model delivery performance of the first model.

[0272] Taking the ratio of the effective time information and the model transfer times as an example, the effective time information can be represented as T0. The model transfer times can be represented as N0. The model transfer performance index information of the first model can be represented as P0. At this time, the model transfer performance index information is positively correlated with the model transfer performance of the first model.

[0273] As shown in FIG. 7, the second node transfers the first model to the first node A and the first node B, and respectively instructs the first node A and the first node B to activate the first model, wherein the effective time information of the first model of the first node A is T0, the effective time information of the first model of the first node B is T0+T1, and T1>0. The effective time information of the first model of the first node B is T0+T1. The model transfer times of the first node A is 2, and the model transfer times of the first node B is 1, therefore, the model transfer performance index information of the first node A is P0+P1, and the model transfer performance index information of the first node B is P0+P1+P2. The model transfer performance index information of the first node B is P0+P1+P2. As can be known from FIG. 7, the model transfer performance index information of the first node A is greater than the model transfer performance index information of the first node B, and the model transfer performance of the first model in the first node A is worse than the model transfer performance of the first model in the first node B.

[0274] In some embodiments, the effective time information can be the duration that the performance of the first model reaches or exceeds the first target performance index, and the effective times information can be the times that the performance of the first model reaches or exceeds the first target performance index. The invalid time information can be the duration that the performance of the first model does not reach or exceed the second target performance index, and the invalid times information can be the times that the performance of the first model does not reach or exceed the second target performance index. Compared with the scheme in the above embodiment that takes the activation time as the effective time, the way of calculating the granularity by taking the satisfaction of the target performance index as the effective time is more accurate, and therefore the determined model transfer performance index information is more accurate.

[0275] Optionally, when there is a special value in the model transmission performance index information, the model transmission information, the first effective state information or the first invalid state information, such as 0 or ∞, the special value can be represented by a special field or other pre-defined manner. Or when the value of the model transmission performance index information, the model transmission information, the first effective state information or the first invalid state information exceeds a preset threshold, such as when the model transmission performance index information has a too small value, it can be represented by a default field. The model transmission performance index information can also be represented by multiple value ranges, such as 0 to 0.1, 0.1 to 0.01, 0.01 to 0.001, and less than 0.001. The model transmission performance index information can also be represented in an explicit or implicit manner, such as directly displaying the value determined by calculation, or implicitly representing the relationship between the calculated model transmission performance index information and the preset model transmission performance index information.

[0276] In some embodiments, when the first model corresponds to a plurality of target performance indicators, each target performance indicator can correspond to a model transmission performance index information respectively, or a third target performance indicator is determined based on the plurality of target performance indicators, and the model transmission performance index information is determined based on the third target performance indicator. Wherein, the determination of the model transmission performance index information corresponding to each target performance indicator can refer to the determination method of the model transmission performance index information in the above scheme.

[0277] In some embodiments, when the first model corresponds to one or more quantization manners, each quantization manner corresponds to a model transmission performance index information respectively.

[0278] Wherein, model quantization refers to converting the character type of the model data parameter to a target character type. Different character types correspond to different numerical precisions and different bit occupancies. That is, the quantization manner of the model can affect the performance and data size of the model.

[0279] For example, the character types include single-precision floating-point number (FP32), half-precision floating-point number (FP16), 8-bit integer (INT8), and 4-bit integer (INT4), etc. FP32 represents a model parameter by 32 bits, including 1 bit of sign bit, 8 bits of exponent bit and 23 bits of decimal bit. FP16 represents a model parameter by 16 bits, including 1 bit of sign bit, 5 bits of exponent bit and 10 bits of decimal bit. The higher the numerical precision corresponding to the character type, the smaller the performance loss of the model, and the higher the performance of the corresponding model, at the same time, the more bits occupied by the model parameter, and the larger the model data.

[0280] In addition to the division according to the character type, the quantization manner can be divided in other forms, for example, divided according to different algorithms, divided according to the execution stage of the quantization operation, etc., which are not limited in the present application.

[0281] Optionally, in the embodiments of the present application, the first node can report the model transmission performance index information corresponding to each quantization manner. The determination manner of the model transmission performance index information can refer to the related content described above, and is not described herein.

[0282] Optionally, in the embodiments of the present application, the model transmission performance index information of the first model is determined by the model transmission performance index information corresponding to one or more quantization manners.

[0283] Taking the ratio of the model transmission times and the valid time information as an example, in the case of the quantization manner Q-1 (for example, INT8), the valid time information is T1, and the model transmission times are N1. In the case of the quantization manner Q-1 (for example, FP32), the valid time information is T2, and the model transmission times are N2. The model transmission performance index information is Wherein, p and q are predefined by the protocol, or determined by negotiation between the terminal and the network device.

[0284] Taking the ratio of the model transmission time and the valid time information as an example, as shown in FIG. 8, the second node transmits the first model to the first node A according to the quantization manner Q-1, and instructs the first node A to activate the first model. Then, the second node transmits the first model to the first node A according to the quantization manner Q-2, and instructs the first node A to activate the first model. Wherein, the model transmission time of the second node transmitting the first model to the first node A according to the quantization manner Q-1 is The model transmission time of the second node transmitting the first model to the first node A according to the quantization manner Q-2 is The valid time information of the first model of the first node A corresponding to the quantization manner Q-1 is The valid time information of the first model of the first node A corresponding to the quantization manner Q-2 is At this time, the model transmission performance index information is At this time, when the weight allocated by the quantization manner Q-2 is larger, the influence on the model transmission performance index information corresponding to the quantization manner Q-2 is greater. For the quantization manner that needs to be focused on, the quantization manner can be allocated with a higher weight, that is, the contribution to the model transmission performance is greater.

[0285] In some embodiments, the first valid state information is the first valid state information determined in at least one first time period; and / or, the first invalid state information is the first invalid state information determined in at least one second time period.

[0286] The first time period and the second time period can be the same time period or different time periods. For different time periods, the first time period and the second time period can have the same length or different lengths. Since the model transfer performance of the first model can be affected by changes in the channel environment, communication configuration adjustment or external physical interference, the model transfer performance changes over time. Therefore, in the embodiments of the present application, the first effective state information and the first invalid state information can be counted in a time segmented manner to further determine the model transfer performance index information of the first model at different times.

[0287] In some embodiments, the at least one first time period includes at least one of:

[0288] at least one time period for performance monitoring of the first model;

[0289] at least one time period in which the first model is in an active state;

[0290] at least one time period in which the first model is in a supportable / applicable / available state;

[0291] at least one time period indicated by the second node;

[0292] or at least one time period determined by the first node;

[0293] and / or, the at least one second time period includes at least one of:

[0294] at least one time period for performance monitoring of the first model;

[0295] at least one time period in which the first model is in an active state;

[0296] at least one time period in which the first model is in a supportable / applicable / available state;

[0297] at least one time period indicated by the second node;

[0298] or at least one time period determined by the first node.

[0299] The above time periods can be predefined or determined by negotiation between the first node and the second node, for example, through RRC configuration. For example, when the time period is default, the time period can be represented and counted by -∞.

[0300] In some embodiments, the at least one first time period is contained in at least one monitoring time window or at least one activation time window or at least one applicable time window or at least one available time window, and the at least one monitoring time window or the at least one activation time window or the at least one applicable time window or the at least one available time window is earlier than a first time unit at which the model transfer performance indicator information is sent; and / or, the at least one second time period is contained in at least one monitoring time window or at least one activation time window or at least one applicable time window or at least one available time window, and the at least one monitoring time window or the at least one activation time window or the at least one applicable time window or the at least one available time window is earlier than a second time unit at which the model transfer performance indicator information is sent.

[0301] The activation time window can be a time range in which the first model is in an activated state. The activation can also be referred to as being configured or enabled. The applicable time window can be a time range in which the first model is in an applicable state. The available time window can be a time range in which the first model is in an available state.

[0302] In the embodiments of the present application, the first time period can also not contain the activation time window or the monitoring time window or the applicable time window or the available time window. For example, the first time unit is not located in the monitoring time window, or the first time unit is not located in the activation time window, or the first time unit is not located in the applicable time window, or the first time unit is not located in the available time window. And / or, the second time unit is not located in the monitoring time window, or the second time unit is not located in the activation time window, or the second time unit is not located in the applicable time window, or the second time unit is not located in the available time window.

[0303] In the embodiments of the present application, the first time period can be earlier than a first time unit at which the model transfer performance indicator information is sent, and the second time period can be earlier than a second time unit at which the model transfer performance indicator information is sent. For example, if the first model is never activated / monitored / in an applicable state / in an available state, the model transfer performance indicator information can be reported according to a default value.

[0304] As shown in FIG. 9, the first time period includes the monitoring time window n and the activation time window M, and the first valid state information is the first valid state information determined in the monitoring time window n and the activation time window M included in the first time period. The first node reports the model transfer performance indicator information after the activation time window M. The first invalid state information is not repeated here.

[0305] For example, the first time unit is located in the monitoring time window, or the first time unit is located in the activation time window. The second time unit is located in the monitoring time window, or the second time unit is located in the activation time window.

[0306] As shown in FIG. 10, the first time period includes a monitoring time window n, a monitoring time window n+1 and an activation time window M, and the first valid state information is the first valid state information determined in the monitoring time window n and the activation time window M included in the first time period. The first node reports the model transmission performance indicator information in the monitoring time window n+1. The first invalid state information is not repeated here.

[0307] The overall flow of the communication method provided in the present application is described above, and the calculation, reporting and statistics of the first information are described below in the case of a terminal, an access node and a core network node / OAM.

[0308] As shown in FIG. 11, the communication method includes the following steps:

[0309] In step 1101, the terminal and the access node determine the calculation manner of the model transmission performance indicator information.

[0310] In some embodiments, step 1101 is an optional step. For example, the calculation manner of the model transmission performance indicator information between the terminal and the access node can be realized by a pre-defined manner. For example, for the calculation manner of the model transmission performance indicator information of multiple quantization manners, the calculation manner can be determined based on the weight corresponding to each quantization manner.

[0311] In step 1102, the access node sends monitoring indication information to the terminal. Correspondingly, the terminal receives the monitoring indication information from the access node.

[0312] The monitoring indication information is used to instruct the terminal to report the model transmission performance indicator information. For example, the monitoring indication information can also indicate the monitoring resource. For example, the monitoring indication information can include the time period T indicated by the access node to the terminal for monitoring the model transmission performance. The monitoring resource can also be indicated by other manners, which is not limited in the present application.

[0313] In some embodiments, step 1102 is an optional step. For example, the terminal reports the model transmission performance indicator information using the capability information, or the terminal reports the model transmission performance indicator information when the model is registered in the network device, or the terminal reports the model transmission performance indicator information when reporting the supported model. The model transmission performance indicator information can be carried in a dedicated information, or can be carried in an RRC message, for example, can be carried in an RRC message carrying the capability information of the terminal device or UAI.

[0314] In step 1103, the terminal and / or the access node perform performance monitoring and calculate the model transmission performance indicator information.

[0315] Optionally, in the case that the model transmission performance index information is determined by the terminal, the terminal can further send the model transmission performance index information to the access node. In the case that the model transmission performance index information is determined by the access node, the access node can further send the model transmission performance index information to the terminal.

[0316] For example, the model transmission performance index information can be actively reported by the terminal, can be requested by the network device to be reported by the terminal, can be actively sent by the network device to the terminal, or can be sent by the network device to the terminal in response to the request of the terminal.

[0317] In some embodiments, the interaction of the model transmission performance index information can occur when the model is requested to be transmitted, or before the model is requested to be transmitted. For example, the terminal can request the network device to transmit the model, while the terminal reports the model transmission performance index information of the requested model, or the network device instructs the terminal to report the model transmission performance index information of the requested model. For another example, the network device requests the terminal to transmit the model, while the network device instructs the terminal to report the model transmission performance index information of the model, or the terminal requests the network device to instruct the terminal to report the model transmission performance index information of the model.

[0318] In some embodiments, the interaction of the model transmission performance index information can occur after the model is requested to be transmitted. For example, after the terminal requests the network device to transmit the model, the terminal reports the model transmission performance index information of the model, or the network device instructs the terminal to report the model transmission performance index information of the model. For another example, after the network device requests the terminal to transmit the model, the network device instructs the terminal to report the model transmission performance index information of the model, or the terminal requests the network device to instruct the terminal to report the model transmission performance index information of the model.

[0319] Step 1104: The access node and / or the terminal instructs the core network node / OAM to update the model transmission performance index information of the first model.

[0320] That is, the model transmission performance index information can be generated by the terminal, or can be generated by the access node; the model transmission performance index information can be statistically counted by the terminal, or can be statistically counted by the access node. For example, in the case that the first model is in an effective state, the access node triggers the deactivation operation of the first model, and the network device can separately count the model transmission performance index information at this time and report it to the core network node or OAM. In the case that the first model is in an effective state, the terminal triggers the deactivation operation of the first model, and the terminal can count the model transmission performance index information at this time, which represents the generalization of the model to the internal conditions of the terminal.

[0321] In some embodiments, the model delivery performance indicator information can be a model attribute at the terminal side, which is suitable for a scenario where the model delivery performance of the terminal is decoupled from the network side vendor. Alternatively, the model delivery performance can be individually counted according to each cell global identifier (CGI). The model delivery performance indicator information reported by the terminal can be the model delivery performance of a single model, or the model delivery performance of the models provided by the terminal vendor to which the terminal belongs.

[0322] In some embodiments, when the time period T for monitoring the model delivery performance is periodic monitoring, the terminal can retain the latest model delivery performance (starting from T or multiple Ts) and update the historical model delivery performance (starting from -∞).

[0323] The above describes the method provided by the present application. In addition, the present application also provides a communication device for implementing the functions described in the above method embodiments.

[0324] It can be understood that, in order to implement the above functions, the communication device comprises a hardware structure and / or a software module for executing each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of the examples described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is implemented in hardware or computer software driven hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0325] The embodiments of the present application can divide the functions of the communication device according to the above method embodiments, for example, each function module can be divided according to each function, or two or more functions can be integrated into one processing module. The integrated module can be implemented in the form of hardware or software function module. It should be noted that the division of modules in the embodiments of the present application is illustrative, and is only a logical function division. When actually implemented, there can be another division manner.

[0326] FIG. 12 shows a structural schematic diagram of a communication device 120. The communication device 120 comprises a processing module 1201 and a transceiver module 1202. The communication device 120 can be used to implement the functions of the first node or the second node described above.

[0327] In some embodiments, the communication device 120 can further comprise a storage module (not shown in FIG. 12) for storing program instructions and data.

[0328] In some embodiments, the transceiver module 1202, which can also be referred to as a transceiver unit, is configured to implement transmit and / or receive functions. The transceiver module 1202 can be constituted by a transceiver circuit, a transceiver, a transceiver, or a communication interface.

[0329] In some embodiments, the transceiver module 1202 can include a receiving module and a transmitting module for performing the receiving and transmitting steps of the method embodiments performed by the first node or the second node, respectively, and / or for supporting other processes of the techniques described herein; and the processing module 1201 can be configured to perform the processing steps of the method embodiments performed by the first node or the second node, and / or for supporting other processes of the techniques described herein.

[0330] When the communication apparatus 120 is configured to implement the functions of the first node:

[0331] The processing module 1201 is configured to acquire first information through the transceiver module 1202; the first information includes model transfer performance indicator information of a first model; the first model is configured to enable an artificial intelligence (AI) / machine learning (ML) function; the model transfer performance indicator information is configured to represent a model transfer performance of the first model; and the transceiver module 1202 is configured to send the first information to a second node.

[0332] In a possible design, the model transfer performance indicator information includes at least one model transfer performance indicator, and the at least one model transfer performance indicator includes at least one model transfer performance indicator value and / or at least one first value range.

[0333] In a possible design, the at least one model transfer performance indicator value includes at least two model transfer performance indicator values, and the at least one first value range includes at least two first value ranges.

[0334] In a possible design, the model transfer performance indicator information includes a mathematical statistic representing the model transfer performance of the first model.

[0335] In a possible design, the model transfer performance indicator information includes a mathematical statistic representing the model transfer performance indicator of the first model.

[0336] In a possible design, the first information includes at least one model transfer performance indicator information; one or more of the at least one model transfer performance indicator information has a corresponding data feature or data feature identifier.

[0337] In a possible design, the at least one model transfer performance indicator information is in one-to-one correspondence with the at least one data feature or data feature identifier.

[0338] In a possible design, the at least one model delivery performance indicator corresponds to the at least one data feature or the data feature identifier.

[0339] In a possible design, the first model corresponds to at least one model delivery performance indicator information, and the at least one model delivery performance indicator information includes first model delivery performance indicator information, where the first model delivery performance indicator information corresponds to the first data feature or the first data feature identifier.

[0340] In a possible design, the at least one model delivery performance indicator information further includes second model delivery performance indicator information, where the second model delivery performance indicator information corresponds to the second data feature or the second data feature identifier.

[0341] In a possible design, the model delivery performance indicator information is determined according to state information and / or model delivery information; the state information includes first active state information and / or first inactive state information.

[0342] In a possible design, the model delivery information includes first model delivery quantization information; the first model delivery quantization information includes model delivery times and / or model delivery time.

[0343] In a possible design, the first active state information corresponds to an active state of the first model, and the active state includes at least one of the following: a state in which performance of the first model reaches or exceeds a first target performance indicator; an activated state of the first model; or, a supportable / suitable / available state of the first model; and / or, the first inactive state information corresponds to an inactive state of the first model, and the inactive state includes at least one of the following: a state in which performance of the first model does not reach or exceed a second target performance indicator; a non-activated state of the first model; or, a non-supportable / non-suitable / non-available state of the first model.

[0344] In a possible design, the performance of the first model reaching or exceeding the first target performance indicator includes that the performance of the first model reaches or exceeds a value of the first target performance indicator, or is within a value range of the first target performance indicator; and / or, the performance of the first model not reaching or exceeding the second target performance indicator includes that the performance of the first model does not reach or exceed a value of the second target performance indicator, or is not within a value range of the second target performance indicator.

[0345] In a possible design, the first effective state information includes first effective quantization information of the first model, the first effective quantization information including effective time information and / or effective number information, where the effective time information includes a length of time during which the first model is in the effective state, and the effective number information includes a number of times during which the first model is in the effective state; and / or, the first ineffective state information includes first ineffective quantization information of the first model, the first ineffective quantization information including ineffective time information and / or ineffective number information, where the ineffective time information includes a length of time during which the first model is in the ineffective state, and the ineffective number information includes a number of times during which the first model is in the ineffective state.

[0346] In a possible design, the model transfer performance indicator information includes at least one of the following: first ratio information, first difference information, and first probability statistical information.

[0347] In a possible design, the first ratio information includes a ratio of the first effective quantization information and the model transfer information, or a ratio of the first ineffective quantization information and the model transfer information, or a ratio of the model transfer information and the first effective quantization information, or a ratio of the model transfer information and the first ineffective quantization information.

[0348] In a possible design, the first effective state information is first effective state information determined in at least one first time period; and / or, the first ineffective state information is first ineffective state information determined in at least one second time period.

[0349] In a possible design, the at least one first time period includes at least one of the following: at least one time period during which performance monitoring is performed on the first model; at least one time period during which the first model is in an activated state; at least one time period during which the first model is in a supportable / applicable / available state; at least one time period indicated by the second node; or, at least one time period determined by the first node; and / or, the at least one second time period includes at least one of the following: at least one time period during which performance monitoring is performed on the first model; at least one time period during which the first model is in an activated state; at least one time period during which the first model is in a supportable / applicable / available state; at least one time period indicated by the second node; or, at least one time period determined by the first node.

[0350] In a possible design, the at least one first time period is included in at least one monitoring time window or at least one activation time window or at least one applicable time window or at least one available time window, and the at least one monitoring time window or the at least one activation time window or the at least one applicable time window or the at least one available time window is earlier than the first time unit at which the model transfer performance indicator is sent; and / or, the at least one second time period is included in at least one monitoring time window or at least one activation time window or at least one applicable time window or at least one available time window, and the at least one monitoring time window or the at least one activation time window or the at least one applicable time window or the at least one available time window is earlier than the second time unit at which the model transfer performance indicator is sent.

[0351] In a possible design, the first time unit is not located in a monitoring time window, or the first time unit is not located in an activation time window, or the first time unit is not located in an applicable time window, or the first time unit is not located in an available time window; and / or, the second time unit is not located in a monitoring time window, or the second time unit is not located in an activation time window, or the second time unit is not located in an applicable time window, or the second time unit is not located in an available time window.

[0352] When the communication apparatus 120 is configured to implement the function of the second node:

[0353] The transceiver module 1202 is configured to receive first information from the first node; the first information includes model transfer performance indicator information of a first model; the first model is configured to enable an artificial intelligence (AI) / machine learning (ML) function; the model transfer performance indicator information is configured to represent model transfer performance of the first model; and the processing module 1201 is configured to determine a model transfer strategy according to the first information.

[0354] In a possible design, the model transfer performance indicator information includes at least one model transfer performance indicator, and the at least one model transfer performance indicator includes at least one model transfer performance indicator value and / or at least one first value range.

[0355] In a possible design, the at least one model transfer performance indicator value includes at least two model transfer performance indicator values, and the at least one first value range includes at least two first value ranges.

[0356] In a possible design, the model transfer performance indicator information includes mathematical statistics configured to represent model transfer performance of the first model.

[0357] In a possible design, the model transfer performance indicator information includes mathematical statistics configured to represent a model transfer performance indicator of the first model.

[0358] In a possible design, the first information includes at least one model delivery performance indicator information; one or more of the at least one model delivery performance indicator information has a corresponding data feature or data feature identifier.

[0359] In a possible design, the at least one model delivery performance indicator information corresponds to the at least one data feature or data feature identifier in a one-to-one manner.

[0360] In a possible design, the at least one model delivery performance indicator corresponds to the at least one data feature or data feature identifier in a one-to-one manner.

[0361] In a possible design, the first model corresponds to at least one model delivery performance indicator information, the at least one model delivery performance indicator information includes first model delivery performance indicator information, and the first model delivery performance indicator information corresponds to a first data feature or a first data feature identifier.

[0362] In a possible design, the at least one model delivery performance indicator information further includes second model delivery performance indicator information, and the second model delivery performance indicator information corresponds to a second data feature or a second data feature identifier.

[0363] In a possible design, the model delivery performance indicator information is determined based on state information and / or model delivery information; the state information includes first active state information and / or first inactive state information.

[0364] In a possible design, the model delivery information includes first model delivery quantization information; the first model delivery quantization information includes a model delivery number and / or a model delivery time.

[0365] In a possible design, the first active state information corresponds to an active state of the first model, and the active state includes at least one of the following: a state in which a performance of the first model reaches or exceeds a first target performance indicator; an activated state of the first model; or, a supportable / suitable / available state of the first model; and / or, the first inactive state information corresponds to an inactive state of the first model, and the inactive state includes at least one of the following: a state in which a performance of the first model does not reach or exceed a second target performance indicator; a non-activated state of the first model; or, a non-supportable / non-suitable / non-available state of the first model.

[0366] In a possible design, the performance of the first model reaching or exceeding the first target performance indicator includes that the performance of the first model reaches or exceeds a value of the first target performance indicator, or is within a value range of the first target performance indicator; and / or, the performance of the first model not reaching or exceeding the second target performance indicator includes that the performance of the first model does not reach or exceed a value of the second target performance indicator, or is not within a value range of the second target performance indicator.

[0367] In a possible design, the first effective state information includes first effective quantization information of the first model, the first effective quantization information including effective time information and / or effective frequency information, where the effective time information includes a length of time during which the first model is in the effective state, and the effective frequency information includes a number of times during which the first model is in the effective state; and / or, the first ineffective state information includes first ineffective quantization information of the first model, the first ineffective quantization information including ineffective time information and / or ineffective frequency information, where the ineffective time information includes a length of time during which the first model is in the ineffective state, and the ineffective frequency information includes a number of times during which the first model is in the ineffective state.

[0368] In a possible design, the model transfer performance indicator information includes at least one of the following: first ratio information, first difference information, and first probability statistical information.

[0369] In a possible design, the first ratio information includes a ratio of the first effective quantization information and the model transfer information, or a ratio of the first ineffective quantization information and the model transfer information, or a ratio of the model transfer information and the first effective quantization information, or a ratio of the model transfer information and the first ineffective quantization information.

[0370] In a possible design, the first effective state information is first effective state information determined in at least one first time period; and / or, the first ineffective state information is first ineffective state information determined in at least one second time period.

[0371] In a possible design, the at least one first time period includes at least one of the following: at least one time period during which performance monitoring is performed on the first model; at least one time period during which the first model is in an activated state; at least one time period during which the first model is in a supportable / applicable / available state; at least one time period indicated by the second node; or, at least one time period determined by the first node; and / or, the at least one second time period includes at least one of the following: at least one time period during which performance monitoring is performed on the first model; at least one time period during which the first model is in an activated state; at least one time period during which the first model is in a supportable / applicable / available state; at least one time period indicated by the second node; or, at least one time period determined by the first node.

[0372] In a possible design, the at least one first time period is included in at least one monitoring time window or at least one activation time window or at least one applicable time window or at least one available time window, and the at least one monitoring time window or the at least one activation time window or the at least one applicable time window or the at least one available time window is earlier than the first time unit at which the model transfer performance indicator is sent; and / or, the at least one second time period is included in at least one monitoring time window or at least one activation time window or at least one applicable time window or at least one available time window, and the at least one monitoring time window or the at least one activation time window or the at least one applicable time window or the at least one available time window is earlier than the second time unit at which the model transfer performance indicator is sent.

[0373] In a possible design, the first time unit is not located in a monitoring time window, or the first time unit is not located in an activation time window, or the first time unit is not located in an applicable time window, or the first time unit is not located in an available time window; and / or, the second time unit is not located in a monitoring time window, or the second time unit is not located in an activation time window, or the second time unit is not located in an applicable time window, or the second time unit is not located in an available time window.

[0374] In the method embodiments, the steps involved in the method embodiments can be implemented by using the corresponding functional modules. Details are not described herein again.

[0375] In this application, the communication apparatus 120 can be in the form of integrated division of various functional modules. The "module" here can refer to a specific application-specific integrated circuit (ASIC), a circuit, a processor and a memory executing one or more software or firmware programs, an integrated logic circuit, and / or other devices that can provide the above functions.

[0376] In some embodiments, when the communication apparatus 120 in FIG. 12 is a chip or a chip system, the functions / implementation procedures of the transceiver module 1202 can be implemented through an input / output interface (or a communication interface) of the chip or the chip system, and the functions / implementation procedures of the processing module 1201 can be implemented through a processor (or a processing circuit) of the chip or the chip system.

[0377] Since the communication apparatus 120 provided in this embodiment can execute the above method, the technical effects that can be achieved by the communication apparatus 120 can refer to the above method embodiments, and details are not described herein again.

[0378] As a possible product form, the first node or the second node described in the embodiments of the present application can be implemented using one or more field programmable gate arrays (FPGAs), programmable logic devices (PLDs), controllers, state machines, gate logic, discrete hardware components, any other suitable circuitry, or any combination of circuitry capable of implementing the various functionalities described throughout the present application.

[0379] As another possible product form, the first node or the second node described in the embodiments of the present application can be implemented by a general bus architecture. For ease of illustration, refer to FIG. 13, which is a structural schematic diagram of a communication apparatus 1300 provided by the embodiments of the present application, the communication apparatus 1300 including a processor 1301 and a transceiver 1302. The communication apparatus 1300 can be the first node, or a chip or chip system therein; or the communication apparatus 1300 can be the second node, or a chip or module therein. FIG. 13 only shows the main components of the communication apparatus 1300. In addition to the processor 1301 and the transceiver 1302, the communication apparatus can further include a memory 1303, and an input / output device (not shown in FIG. 13).

[0380] Optionally, the processor 1301 is mainly used for processing communication protocols and communication data, and controlling the entire communication apparatus, executing software programs, processing data of the software programs, so as to implement the methods provided in the method embodiments described above. The memory 1303 is mainly used for storing software programs and data. The transceiver 1302 can include radio frequency circuitry and an antenna, the radio frequency circuitry being mainly used for conversion between baseband signals and radio frequency signals, and processing of the radio frequency signals. The antenna is mainly used for transceiving radio frequency signals in the form of electromagnetic waves. The input / output device, such as a touch screen, a display screen, a keyboard, etc., is mainly used for receiving data input by a user and outputting data to the user.

[0381] Optionally, the processor 1301, the transceiver 1302, and the memory 1303 can be connected through a communication bus.

[0382] When the communication apparatus is powered on, the processor 1301 can read the software program in the memory 1303, execute the instructions of the software program, and process the data of the software program. When data needs to be transmitted wirelessly, the processor 1301 performs baseband processing on the data to be transmitted, and outputs the baseband signal to the radio frequency circuit. The radio frequency circuit performs radio frequency processing on the baseband signal, and transmits the radio frequency signal in the form of electromagnetic wave through the antenna. When data is transmitted to the communication apparatus, the radio frequency circuit receives the radio frequency signal through the antenna, converts the radio frequency signal into a baseband signal, and outputs the baseband signal to the processor 1301. The processor 1301 converts the baseband signal into data and processes the data.

[0383] In another implementation, the radio frequency circuit and the antenna can be arranged independently of the processor performing the baseband processing, for example, in a distributed scenario, the radio frequency circuit and the antenna can be arranged remotely from the communication apparatus.

[0384] In some embodiments, in a hardware implementation, those skilled in the art can conceive that the above-mentioned communication apparatus 120 can adopt the form of the communication apparatus 1300 shown in FIG. 13.

[0385] As an example, the functions / implementation processes of the processing module 1201 in FIG. 12 can be implemented by the processor 1301 in the communication apparatus 1300 shown in FIG. 13 invoking the computer-executable instructions stored in the memory 1303. The functions / implementation processes of the transceiver module 1202 in FIG. 12 can be implemented by the transceiver 1302 in the communication apparatus 1300 shown in FIG. 13.

[0386] As another possible product form, the first node or the second node in the present application can adopt the constituent structure shown in FIG. 14, or include the components shown in FIG. 14. FIG. 14 is a constituent diagram of a communication apparatus 1400 provided in the present application. The communication apparatus 1400 can be the first node or a chip or system on chip in the first node; or can be the second node or a chip or system on chip in the second node.

[0387] As shown in FIG. 14, the communication apparatus 1400 includes at least one processor 1401, and at least one communication interface (only one communication interface 1404 is shown in FIG. 14 as an example, and the processor 1401 is taken as an example for description). Optionally, the communication apparatus 1400 can further include a communication bus 1402 and a memory 1403.

[0388] The processor 1401 can be a general-purpose central processing unit (CPU), a general-purpose processor, a network processing unit (NP), a digital signal processing (DSP), a microprocessor, a microcontroller, a PLD, or any combination thereof. The processor 1401 can also be other apparatuses with processing capabilities, such as a circuit, a device, or a software module, without limitation.

[0389] The communication bus 1402 is used to connect different components in the communication apparatus 1400, so that different components can communicate. The communication bus 1402 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is shown in FIG. 14, but it does not mean that there is only one bus or only one type of bus.

[0390] The communication interface 1404 is used to communicate with other devices or communication networks. For example, the communication interface 1404 can be a module, a circuit, a transceiver, or any device capable of communication. Alternatively, the communication interface 1404 can also be an input / output interface in the processor 1401, used to realize the signal input and signal output of the processor.

[0391] The memory 1403 can be a device with a storage function, used to store instructions and / or data. The instructions can be a computer program.

[0392] For example, the memory 1403 can be a read-only memory (ROM) or other type of static storage device that can store static information and / or instructions, or a random access memory (RAM) or other type of dynamic storage device that can store information and / or instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM), or other optical disk storage, optical disk storage (including compact disks, laser disks, optical disks, digital versatile disks, Blu-ray disks, etc.), magnetic disk storage media, or other magnetic storage devices, etc., without limitation.

[0393] It should be noted that the memory 1403 can exist independently of the processor 1401, or can be integrated into the processor 1401. The memory 1403 can be located within the communication device 1400, or can be located outside the communication device 1400, without limitation. The processor 1401 can be configured to execute instructions stored in the memory 1403 to implement the methods provided by the embodiments described below.

[0394] Optionally, the processor 1401 and / or the memory 1403 can include an artificial intelligence (AI) module, which is configured to implement AI-related functions. The AI module can be implemented in software, hardware, or a combination of software and hardware. For example, the AI module can include a radio access network intelligent controller (RIC) module. For example, the AI module can be a near-real-time RIC or a non-real-time RIC.

[0395] As an optional implementation, the communication device 1400 can further include an output device 1405 and an input device 1406. The output device 1405 is in communication with the processor 1401 and can display information in various ways. For example, the output device 1405 can be a liquid crystal display (LCD), a light emitting diode (LED) display device, a cathode ray tube (CRT) display device, or a projector, etc. The input device 1406 is in communication with the processor 1401 and can receive user input in various ways. For example, the input device 1406 can be a mouse, a keyboard, a touch screen device, a sensor device, etc.

[0396] In some embodiments, in a hardware implementation, those skilled in the art can conceive that the communication device 120 shown in FIG. 12 can take the form of the communication device 1400 shown in FIG. 14.

[0397] As an example, the functions / implementation processes of the processing module 1201 in FIG. 12 can be implemented by the processor 1401 in the communication device 1400 in FIG. 14 invoking computer execution instructions stored in the memory 1403. The functions / implementation processes of the transceiver module 1202 in FIG. 12 can be implemented by the communication interface 1404 in the communication device 1400 in FIG. 14.

[0398] It should be noted that the structure shown in FIG. 14 does not constitute a specific limitation on the first node or the second node. For example, in some embodiments of the present application, the first node or the second node can include more or fewer components than shown, or combine certain components, or split certain components, or different arrangement of components. The components shown can be implemented in hardware, software, or a combination of software and hardware.

[0399] In some embodiments, the present application also provides a communication apparatus, which includes a processor for implementing the method in any of the above method embodiments.

[0400] As a possible implementation, the communication apparatus further includes a memory. The memory is used to store necessary computer programs and data. The computer programs can include instructions, and the processor can invoke the instructions in the computer programs stored in the memory to instruct the communication apparatus to perform the method in any of the above method embodiments. Of course, the memory can also not be in the communication apparatus.

[0401] As another possible implementation, the communication apparatus further includes an interface circuit, which is a code / data read-write interface circuit, and is used to receive computer execution instructions (computer execution instructions are stored in the memory, which can be directly read from the memory or can pass through other devices) and transmit to the processor.

[0402] As yet another possible implementation, the communication apparatus further includes a communication interface, which is used to communicate with modules outside the communication apparatus.

[0403] It can be understood that the communication apparatus can be a chip or a chip system. When the communication apparatus is a chip system, it can be composed of a chip or include a chip and other discrete devices, and the embodiments of the present application do not make specific limitations.

[0404] The present application also provides a computer readable storage medium, which stores computer programs or instructions, and the computer programs or instructions are executed by a computer to realize the functions of any of the above method embodiments.

[0405] The present application also provides a computer program product, which is executed by a computer to realize the functions of any of the above method embodiments.

[0406] Those skilled in the art can understand that, for the convenience and brevity of description, the specific working processes of the above-described system, apparatus and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0407] It can be understood that the system, apparatus and method described in the present application can also be implemented in other manners. For example, the apparatus embodiment described above is merely illustrative. For example, the division of the units is only a logical function division. There can be another division manner for the actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0408] The units described as separate components can or can not be physically separate, i.e., can be located in one place, or can be distributed on a plurality of network units. The components shown as units can or can not be physical units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0409] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can be physically present separately, or two or more units can be integrated into one unit.

[0410] In the above embodiments, all or part can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions can be stored in a computer readable storage medium or transferred from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transferred from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) manner. The computer readable storage medium can be any available medium that can be accessed by a computer or include one or more data storage devices such as servers, data centers, etc. integrated with the medium. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state drive (SSD)), etc. In the embodiments of the present application, the computer can include the apparatus described above.

[0411] Although the application has been described in connection with various embodiments thereof, it will be understood that other modifications and variations will be apparent to those skilled in the art in view of the foregoing disclosure, the drawings, and the accompanying claims. It is therefore contemplated that the application will be practiced otherwise than as specifically set forth herein. For example, claims can be presented that are broader in scope than the above described embodiments. Accordingly, the specification and drawings are to be regarded in an illustrative, rather than a restrictive, sense. The disclosure covers any and all modifications, variations, combinations or equivalents that fall within the scope of the present application. It is contemplated that the application will be practiced in the absence of any element of the application not specifically disclosed herein. It is also contemplated that one or more claimed elements can be invoked in the claim depending on the requirements of the application.

[0412] Although the application has been described in connection with specific embodiments thereof, it will be understood that various modifications and variations will be apparent to those skilled in the art in view of the foregoing disclosure, the drawings, and the accompanying claims. Accordingly, it is intended that the application be covered by all such modifications and variations that fall within the scope of the application. It is also intended that the application encompass all alternatives, modifications and equivalents falling within the scope of the claims. Obviously, many modifications and variations of the present application are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

A communication method characterized by comprising: The method applied to a first node comprises: obtaining first information; the first information comprises model delivery performance indicator information of a first model; wherein the first model is used to enable an artificial intelligence (AI) / machine learning (ML) function; and the model delivery performance indicator information is used to represent model delivery performance of the first model; sending the first information to a second node. A communication method characterized by comprising: The method applied to a second node comprises: receiving first information from a first node; the first information comprises model delivery performance indicator information of a first model; wherein the first model is used to enable an artificial intelligence (AI) / machine learning (ML) function; and the model delivery performance indicator information is used to represent model delivery performance of the first model; determining a model delivery strategy according to the first information. The method according to claim 1 or 2, characterized in that The model delivery performance indicator information comprises at least one model delivery performance indicator, and the at least one model delivery performance indicator comprises at least one model delivery performance indicator value and / or at least one first value range. The method according to any one of claims 1 to 3, characterized in that The model delivery performance indicator information comprises mathematical statistics used to represent model delivery performance of the first model. The method according to any one of claims 1 to 4, characterized in that The first information comprises at least one model delivery performance indicator information; one or more model delivery performance indicator information in the at least one model delivery performance indicator information has a corresponding data feature or data feature identifier. The method according to any one of claims 1 to 5, characterized in that The model delivery performance indicator information is determined by state information and / or model delivery information; the state information comprises first effective state information and / or first invalid state information. The method according to claim 6, characterized in that The model delivery information comprises first model delivery quantitative information; the first model delivery quantitative information comprises model delivery times and / or model delivery time. The method according to claim 6 or 7, characterized in that The first effective state information corresponds to an effective state of the first model, and the effective state comprises at least one of the following: a state in which performance of the first model reaches or exceeds a first target performance indicator; an activated state of the first model; or a state in which the first model is supportable / suitable / available. And / or, the first invalid state information corresponds to an invalid state of the first model, and the invalid state comprises at least one of the following: a state in which performance of the first model does not reach or exceed a second target performance indicator; a non-activated state of the first model; or a state in which the first model is non-supportable / non-suitable / non-available. The first effective state information comprises first effective quantitative information of the first model, and the first effective quantitative information comprises effective time information and / or effective times information, wherein the effective time information comprises a length of time during which the first model is in an effective state, and the effective times information comprises a number of times during which the first model is in an effective state. The method according to any one of claims 6-8, characterized in that And / or, the first invalid state information comprises first invalid quantitative information of the first model, and the first invalid quantitative information comprises invalid time information and / or invalid times information, wherein the invalid time information comprises a length of time during which the first model is in an invalid state, and the invalid times information comprises a number of times during which the first model is in an invalid state. ​ The method according to any one of claims 6-9, characterized in that The first effective state information is first effective state information determined in at least one first time period; And / or, the first invalid state information is first invalid state information determined in at least one second time period. The method of claim 10, wherein The at least one first time period comprises at least one of the following: At least one time period in which performance monitoring is performed on the first model; At least one time period in which the first model is in an activated state; At least one time period in which the first model is in a supportable / applicable / available state; At least one time period indicated by the second node; Or, at least one time period determined by the first node; And / or, the at least one second time period comprises at least one of the following: At least one time period in which performance monitoring is performed on the first model; At least one time period in which the first model is in an activated state; At least one time period in which the first model is in a supportable / applicable / available state; At least one time period indicated by the second node; Or, at least one time period determined by the first node. The method according to claim 10 or 11, characterized in that The at least one first time period is contained in at least one monitoring time window or at least one activated time window or at least one applicable time window or at least one available time window, the at least one monitoring time window or the at least one activated time window or the at least one applicable time window or the at least one available time window being earlier than a first time unit in which the model transfer performance indicator information is sent; And / or, the at least one second time period is contained in at least one monitoring time window or at least one activated time window or at least one applicable time window or at least one available time window, the at least one monitoring time window or the at least one activated time window or the at least one applicable time window or the at least one available time window being earlier than a second time unit in which the model transfer performance indicator information is sent. A communication device characterized by comprising: Comprise: A functional unit for performing the method of any one of claims 1-12; wherein the actions performed by the functional unit are implemented by hardware or corresponding software executed by hardware. A communication device characterized by comprising: The communication device comprises a processor; the processor is used to run computer programs or instructions, so that the communication device performs the method of any one of claims 1-12. A computer-readable storage medium, characterized by The computer readable storage medium stores computer instructions or programs, when the computer instructions or programs are run on the computer, so that the method of any one of claims 1-12 is executed. A computer program product, characterized in that The computer program product comprises computer instructions; when part or all of the computer instructions are run on the computer, so that the method of any one of claims 1-12 is executed.

Images