Communication method and device

Abstract

The present application provides a communication method and device. The method comprises: receiving first information, wherein the first information is used for instructing to simulate the capability of AI / ML and / or to simulate the inference type of the AI / ML; performing simulation on the AI / ML on the basis of the first information to obtain a simulation report; and sending the simulation report. A first device can determine a simulation report on the basis of the first information and send the simulation report, so that a device receiving the simulation report can perform corresponding processing on the basis of the simulation report, thereby improving the simulation effect. In addition, the first information can instruct to stimulate the capability and / or the inference type of the AI / ML, guiding the simulation behavior of the first device in a more fine-grained manner, thereby improving the simulation effect.

Description

Communication method and device

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office of China on January 16, 2024, with application number 202410063778.0 and application name “Communication Method and Device”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of communication technology, and more particularly, to a communication method and apparatus. Background Art

[0003] Artificial intelligence (AI) and machine learning (ML) technologies are increasingly being adopted by a wider range of industries. th In the 5GS (5th generation system), AI / ML functions can be applied to intelligent optimization use cases in wireless access networks, management data analysis services in network management, and network data analysis services in core networks.

[0004] The AI / ML workflow can include training, simulation, model deployment, and inference. During the simulation phase, AI / ML entities used for inference can be run in a simulation environment to evaluate their inference performance and prevent unsatisfactory AI / ML entities from being deployed in the next phase.

[0005] However, how to improve the simulation effect of AI / ML is an urgent problem to be solved. Summary of the Invention

[0006] The present application provides a communication method and device that can improve the simulation effect of AI / ML.

[0007] In a first aspect, a communication method is provided, comprising: receiving first information, wherein the first information is used to indicate simulation of AI / ML capabilities and / or simulation of the reasoning type of the AI / ML; performing simulation on the AI / ML based on the first information to obtain a simulation report; and sending the simulation report.

[0008] Optionally, the method may be performed by a first device, or by a component in the first device (e.g., a processor, a chip, or a chip system), or by a logic module or software that implements all or part of the functions of the first device. Exemplarily, the first device may be a domain management function (Domain-MnF), such as an element management system (EMS), a mobile broadband automation engine (MAE), a management service (MnS) consumer, a management data analytics (MDA) producer, or other Domain-MnFs.

[0009] Through the above embodiment, the first device can perform simulation according to the first information and send a simulation report, so that the device receiving the simulation report can make corresponding processing according to the simulation report, thereby improving the simulation effect. In addition, the first information can indicate the simulation of the capabilities and / or reasoning types of AI / ML. Among them, the capabilities of AI / ML can refer to the analysis capabilities of AI / ML for one or more indicators, and the reasoning type of AI / ML can refer to the ability of AI / ML to reason about a certain type of task. Through the first information, the simulation behavior of the first device can be indicated in a more fine-grained manner, thereby improving the simulation effect.

[0010] In combination with the first aspect, in certain implementations of the first aspect, the first information also includes at least one of the following: a model identifier of the AI / ML; a model identifier group of the AI / ML; an activation status, used to indicate whether to turn on or off the simulation of the AI / ML; a simulation area, used to indicate the area where the AI / ML is simulated; a simulation time, used to indicate the time when the AI / ML is simulated; a first request, used to request the simulation result of the capability of the AI / ML; a second request, used to request the simulation result of the reasoning type; cycle information, used to indicate the sending cycle of the simulation report; a performance indicator of the capability of the AI / ML; a performance indicator of the reasoning type of the AI / ML model; a simulation strategy, including at least one of the performance monitoring threshold, performance monitoring area or performance monitoring time of the AI / ML model; or a termination strategy, used to indicate the conditions for terminating the simulation of the AI / ML.

[0011] Through the above embodiment, the first information contains more information, thereby more precisely indicating the simulation behavior of the first device, thereby further improving the simulation effect.

[0012] In combination with the first aspect, in certain implementations of the first aspect, the AI / ML capabilities include at least one of the following: traffic analysis, coverage analysis, mobility analysis, load analysis, fault analysis, service experience analysis, energy efficiency analysis, or energy consumption analysis.

[0013] Through the above embodiments, the first information can instruct the first device to simulate at least one of the ML capabilities of traffic, coverage, mobility, load, fault, service experience, energy efficiency or energy consumption, thereby more finely indicating the simulation behavior of the first device and further improving the simulation effect, so as to evaluate the reasoning performance of the ML entity corresponding to the ML capability of traffic, coverage, mobility, load, fault, service experience, energy efficiency or energy consumption in the simulation environment before applying it to the destination network or system.

[0014] In combination with the first aspect, in certain implementations of the first aspect, the reasoning type of the AI / ML includes at least one of the following: an reasoning type of MDA, an reasoning type of a self-organizing network (SON), an reasoning type of a network data analysis function, or an reasoning type of radio access network intelligence (RAN intelligence).

[0015] Through the above embodiments, the first information can instruct the first device to analyze at least one reasoning type among the management data analysis function, the self-optimizing network function, the network data analysis function, or the wireless access network intelligent function, thereby more finely indicating the simulation behavior of the first device, thereby further improving the simulation effect, so as to evaluate the performance of the management data analysis function, the self-optimizing network function, the network data analysis function, or the wireless access network intelligent function in the simulation environment before applying it to the target network or system.

[0016] In combination with the first aspect, in certain implementations of the first aspect, the simulation report includes at least one of the following: a model identifier of the AI / ML; a model identifier group of the AI / ML; a simulation status, used to indicate the status of simulating the AI / ML; a simulation reasoning capability result, used to indicate the simulation result of the capability of the AI / ML; or a simulation reasoning type result, used to indicate the simulation result of the reasoning type of the AI / ML.

[0017] Through the above embodiments, the simulation report contains more information, so that the device receiving the simulation report can make corresponding processing, thereby further improving the simulation effect.

[0018] In combination with the first aspect, in some implementations of the first aspect, the simulation reasoning capability result includes at least one of the following: traffic congestion recovery suggestions, cell / area average throughput, edge throughput, traffic congestion recovery suggestions, weak coverage, over-coverage, coverage holes, cross-area coverage, coverage problem area information, coverage problem cell identification, reference signal received power (RSRP) distribution, signal to interference plus noise ratio (SINR) distribution, reference signal received quality (RSRQ) distribution, coverage optimization suggestions, switching success rate, switching failure rate, number of early switches, number of late switches, number of switches to wrong cells, mobility optimization suggestions, cell physical resource block (PRB) utilization, number of cell user connections, central processing unit (CPU) unit, CPU) utilization, load optimization suggestions, fault problem, fault problem cell identifier, fault severity, fault recovery suggestions, energy efficiency problem, energy efficiency problem cell or base station identifier, energy consumption problem, energy consumption problem cell or base station identifier, available output power of the base station, number of users served at a specific power, or energy saving suggestions.

[0019] Through the above embodiments, the simulation report may include one or more performance indicators of traffic analysis, coverage analysis, mobility analysis, load analysis, fault analysis, service experience analysis, energy efficiency analysis or energy consumption analysis, so that the device receiving the simulation report can make corresponding processing according to the various indicators of AI / ML capabilities, thereby further improving the simulation effect.

[0020] In combination with the first aspect, in certain implementations of the first aspect, the simulation results of the AI / ML reasoning type include at least one of the following: an identifier of a cell / base station, information of a shut-down cell / carrier / time slot, a cell individual offset (CIO), a trigger time (TTT), or a performance indicator of the reasoning type.

[0021] Through the above embodiments, the simulation report may include at least one simulation result and / or performance indicator of the reasoning type, so that the device receiving the simulation report can make corresponding processing according to the simulation result or various indicators of the AI / ML reasoning type, thereby further improving the simulation effect.

[0022] In a second aspect, a communication method is provided, comprising: sending first information, wherein the first information is used to indicate simulation of AI / ML capabilities and / or simulation of the reasoning type of the AI / ML; and receiving a simulation report.

[0023] Optionally, the method may be performed by a second device, or by a component in the second device (e.g., a processor, a chip, or a chip system), or by a logic module or software that implements all or part of the functions of the second device. Exemplarily, the second device may be a cross-domain management function (CD-MnF), such as a network management system (NMS), an MnS producer, an MDA producer, or other CD-MnF.

[0024] In combination with the second aspect, in certain implementations of the second aspect, the first information also includes at least one of the following: a model identifier of the AI / ML; a model identifier group of the AI / ML; an activation status, used to indicate whether to turn on or off the simulation of the AI / ML; a simulation area, used to indicate the area where the AI / ML is simulated; a simulation time, used to indicate the time when the AI / ML is simulated; a first request, used to request the simulation result of the AI / ML capability; a second request, used to request the simulation result of the reasoning type; cycle information, used to indicate the sending cycle of the simulation report; performance indicators of the AI / ML capability; performance indicators of the reasoning type of the AI / ML model; a simulation strategy, including at least one of the performance monitoring threshold, performance monitoring area or performance monitoring time of the AI / ML model; or a termination strategy, used to indicate the conditions for terminating the simulation of the AI / ML.

[0025] In conjunction with the second aspect, in certain implementations of the second aspect, the AI / ML capability includes at least one of the following:

[0026] Capabilities of traffic analysis, coverage analysis, mobility analysis, load analysis, fault analysis, service experience analysis, energy efficiency analysis, or energy consumption analysis.

[0027] In combination with the second aspect, in certain implementations of the second aspect, the reasoning type of the AI / ML includes at least one of the following: the reasoning type of MDA, the reasoning type of SON, the reasoning type of network data analysis function, or the reasoning type of wireless access network intelligence.

[0028] In combination with the second aspect, in certain implementations of the second aspect, the simulation report includes at least one of the following: a model identifier of the AI / ML; a model identifier group of the AI / ML; a simulation status, used to indicate the status of simulating the AI / ML; a simulation reasoning capability result, used to indicate the simulation result of the capability of the AI / ML; or a simulation reasoning type result, used to indicate the simulation result of the reasoning type of the AI / ML.

[0029] In combination with the second aspect, in certain implementations of the second aspect, the simulation reasoning capability results include at least one of the following: traffic congestion problems, burst traffic congestion, non-burst traffic congestion, cell / area average throughput, edge throughput, traffic congestion recovery suggestions, weak coverage, over-coverage, coverage holes, cross-area coverage, coverage problem area information, coverage problem cell identification, RSRP distribution, SINR distribution, RSRQ distribution, coverage optimization suggestions, switching success rate, switching failure rate, number of early switches, number of late switches, number of switches to wrong cells, mobility optimization suggestions, cell PRB utilization, number of cell user connections, CPU usage, load optimization suggestions, fault problems, fault problem cell identification, fault severity, fault recovery suggestions, energy efficiency problems, energy efficiency problem cell or base station identification, energy consumption problems, energy consumption problem cell or base station identification, available output power of the base station, number of users served at a specific power, or energy-saving suggestions.

[0030] In combination with the second aspect, in certain implementations of the second aspect, the simulation results of the AI / ML reasoning type include at least one of the following: an identifier of a cell / base station, information of a shut-down cell / carrier / time slot, CIO, TTT, or a performance indicator of the reasoning type.

[0031] In a third aspect, a communication device is provided, comprising a processor, wherein the processor is configured to enable the communication device to execute the first aspect and any possible method of the first aspect, or enable the communication device to execute the first aspect and any possible method of the second aspect, by executing a computer program or instruction, or by processing a circuit.

[0032] In one possible implementation, the communication device further includes a memory for storing the computer program or instruction. Furthermore, the processor is specifically configured to call and execute the computer program or computer instruction stored in the memory, so that the processor implements any one of the implementations of the first aspect or the second aspect.

[0033] In one possible implementation, the communication device further includes a transceiver (also referred to as a communication interface), the transceiver being configured to input and / or output signals via the communication interface, and the processor being configured to control the transceiver to transmit and receive signals.

[0034] In a fourth aspect, a communication device is provided, comprising a processing circuit and an input / output interface (also referred to as an interface circuit), the input / output interface being used to input and / or output signals, the processing circuit being used to execute the first aspect and any possible method of the first aspect; or the processing circuit being used to execute the second aspect and any possible method of the second aspect.

[0035] In one possible implementation, the processor is configured to communicate with other devices via an interface circuit and execute any one of the implementations in the first aspect or any one of the implementations in the second aspect.

[0036] In a fifth aspect, a communication device is provided. The communication device may be a first device, or a device or module for performing the function of the first device; the communication device may be a second device, or a device or module for performing the function of the second device.

[0037] In one possible implementation, the communication device may include a module or unit corresponding to each of the methods / operations / steps / actions described in the first aspect. The module or unit may be a hardware circuit, software, or a combination of hardware circuit and software.

[0038] In another possible implementation, the communication device may include a module or unit corresponding to each of the methods / operations / steps / actions described in the second aspect. The module or unit may be a hardware circuit, software, or a combination of hardware circuit and software.

[0039] In the sixth aspect, a computer-readable storage medium is provided, on which a computer program or instruction is stored. When the computer program or the instruction is run on a computer, the first aspect and any possible method of the first aspect are executed; or, the second aspect and any possible method of the second aspect are executed.

[0040] In the seventh aspect, a computer program product is provided, comprising a computer program or instructions, which, when run on a computer, causes the first aspect and any possible method of the first aspect to be executed; or causes the second aspect and any possible method of the second aspect to be executed.

[0041] In an eighth aspect, a communication device is provided, comprising a processor connected to a memory and configured to call a program stored in the memory to execute any possible method of the first aspect or any possible method of the second aspect. The memory may be located within or outside the communication device. The processor may include one or more processors.

[0042] In one implementation, the communication device of the third, fourth, fifth, and eighth aspects above may be a chip or a chip system. In the case where the communication device is a chip or a chip system, the "sending" and "receiving" actions of the communication device may be understood as the chip or chip system sending or receiving to other devices or units (e.g., communication interfaces) in the device including the chip or chip system through pins.

[0043] In a ninth aspect, a chip device is provided, comprising a processor for calling a computer program or computer instruction in a memory so that the processor executes any one of the implementations in the first aspect or any one of the implementations in the second aspect.

[0044] Optionally, the processor is coupled to the memory via an interface.

[0045] In a tenth aspect, a communication system is provided, which includes a first device and a second device; the first device is used to execute the method shown in the first aspect, and the second device is used to execute the method shown in the second aspect.

[0046] In the eleventh aspect, a communication method is provided, which is applied to a first device and a second device, wherein the method includes: the first device executes the method shown in the first aspect; the second device executes the method shown in the second aspect.

[0047] The description of the advantageous effects of any of the second aspect to the eleventh aspect etc. may refer to the description of the advantageous effects of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] FIG1 is a schematic diagram of a network architecture of a communication system applicable to an embodiment of the present application.

[0049] FIG2 is a schematic diagram of a network architecture of another communication system applicable to an embodiment of the present application.

[0050] FIG3 is a schematic flowchart of a communication method provided in an embodiment of the present application.

[0051] FIG4 is a schematic diagram of an implementation model of an AI / ML simulation provided in this application.

[0052] FIG5 is a schematic block diagram of a communication device according to an embodiment of the present application.

[0053] FIG6 is a schematic block diagram of another communication device according to an embodiment of the present application.

[0054] FIG7 is a schematic block diagram of a communication system according to an embodiment of the present application. DETAILED DESCRIPTION

[0055] The technical solution in this application will be described below with reference to the accompanying drawings.

[0056] The technical solutions provided in this application can be applied to various communication systems, such as 5G or new radio (NR) systems, long term evolution (LTE) systems, LTE frequency division duplex (FDD) systems, LTE time division duplex (TDD) systems, etc. The technical solutions provided in this application can also be applied to future communication systems, such as the sixth generation (6G) th The technical solution provided in this application can also be applied to device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-to-machine (M2M) communication, machine type communication (MTC), and Internet of Things (IoT) communication systems or other communication systems.

[0057] Figure 1 is a schematic diagram of the network architecture of a communication system applicable to an embodiment of the present application. The communication system may adopt a service-oriented management architecture. The communication system may include a business support system (BSS), CD-MnF, Domain-MnF, and network elements.

[0058] A BSS can be a system for communication services. It can provide functions and management services such as billing, settlement, accounting, customer service, operations, network monitoring, communication service lifecycle management, and service intent translation. For example, a BSS can be a carrier's operational system, a vertical operational technology system, or other systems.

[0059] The CD-MnF can provide one or more of the following MnF or MnS functions: network lifecycle management, network deployment, network fault management, network performance management, network configuration management, network assurance, network optimization, and translation of the network intent from the communication service provider (Intent-CSP). The CD-MnF can also be referred to as a network management function (NMF). For example, the CD-MnF can be a node (or network management entity) such as an NMS, an MnS producer, an MnS consumer, or a network function management service consumer (NFMS_C).

[0060] The network referred to in the above MnF or MnS may include one or more network elements or subnetworks, or a network slice. That is, NMF can be a network slice management function unit (NSMF), or a cross-domain management data analytical function unit (MDAF), or a cross-domain self-organization network function (SON Function) or a cross-domain intent driven management service (Intent Driven MnS).

[0061] Optionally, in certain deployment scenarios, CD-MnF can also provide subnetwork lifecycle management, subnetwork deployment, subnetwork fault management, subnetwork performance management, subnetwork configuration management, subnetwork assurance, subnetwork optimization, and translation of the network intent of the subnetwork service producer (Intent-CSP) or the network intent of the subnetwork service consumer (Intent-CSC). The subnetwork here consists of multiple small subnetworks and can be a network slice subnetwork.

[0062] Domain-MnF can provide one or more of the following MnF or MnS: lifecycle management of sub-networks or network elements, deployment of sub-networks or network elements, fault management of sub-networks or network elements, performance management of sub-networks or network elements, assurance of sub-networks or network elements, optimization functions of sub-networks or network elements, and translation of intent (intent from network operator, Intent-NOP) of sub-networks or network elements, etc. The sub-network here includes one or more network elements. A sub-network may also include a sub-network, that is, one or more sub-networks form a larger sub-network. Domain-MnF can also be called NMF or network element management functional unit. For example, Domain-MnF can be a node (or network element management entity) such as EMS, MAE, MnS producer, MnS consumer or network function management service provider (NFMS_P).

[0063] Optionally, the subnetwork here can also be a network slice subnetwork. The domain management system can be a network slice subnet management function unit (NSSMF), a domain management data analytical function unit (Domain MDAF), a domain self-organization network function (SON Function), a domain intent management function unit, etc.

[0064] Among them, Domain-MnF can be classified as follows.

[0065] By network type, they can be divided into: radio access network domain management function (RAN-Domain-MnF), core network domain management function (CN-Domain-MnF), transport network domain management function (TN-Domain-MnF), etc. It should be noted that Domain-MnF can also be a domain network management system that can manage one or more of the access network, core network, or transport network.

[0066] According to the administrative region classification, it can be divided into: domain management functional units of a certain area, such as the domain management functional units of City A, etc.

[0067] It should be noted that a CD-MnF can be either a MnS producer or a MnS consumer; a Domain-MnF can be either a MnS producer or a MnS consumer. For example, if MnS is provided by a CD-MnF, the CD-MnF is the MnS producer and the BSS is the MnS consumer. For another example, if MnS is provided by a Domain-MnF, the Domain-MnF is the MnS producer and the CD-MnF is the MnS consumer. For another example, if MnS is provided by a network element, the network element is the MnS producer and the Domain-MnF is the MnS consumer.

[0068] A network element may be an entity that provides network services. A network element may provide one or more of the following management functions or management services: lifecycle management of network elements, deployment of network elements, fault management of network elements, performance management of network elements, assurance of network elements, optimization functions of network elements, and translation of network element intent, etc. Network elements may include core network elements, access network elements, etc. Exemplarily, core network elements may include: access and mobility management function (AMF), session management function (SMF), policy control function (PCF), network data analytical function (NWDAF), network repository function (NRF), and network functions (NF) such as gateways. Exemplarily, the access network elements may include: base stations (such as next generation node basestations (gNBs), evolved node Bs (eNBs or eNodeBs), etc.), centralized control units (central unit control planes, CUCPs), centralized units (central units, CUs), distributed units (distribution units, DUs), centralized user plane units (central unit user planes, CUUPs), radio network controllers (radio network controllers, RNCs), etc. Among them, the RNC may also be called a base station controller.

[0069] Figure 2 is a schematic diagram of the network architecture of another communication system applicable to an embodiment of the present application. The network architecture shown in Figure 2 can be applied to the NR system.

[0070] Referring to Figure 2, MnF can be a 3GPP (3 rd A management entity defined by the 3rd Generation Partnership Project (3GPP). The externally visible behavior and interfaces of a MnF are defined as MnS. In a service-providing management architecture, an MnF can be either a MnS producer or a MnS consumer. When an MnF acts as an MnS producer, the MnS produced by the MnF may have multiple consumers. An MnF can consume multiple MnSs from one or more MnS producers. In other words, as an MnS consumer, an MnF can consume MnSs produced by multiple MnS producers.

[0071] As shown in Figure 2, an MnF can function as both a MnS producer and a MnS consumer. In the example shown in Figure 2, the MnF, acting as a producer, provides MnS to two MnS consumers. Simultaneously, the MnF, acting as a consumer, consumes MnS produced by three MnS producers. It should be noted that Figure 2 is merely an example. This application does not limit the amount of MnS that a MnF can produce or consume; the MnF may produce or consume more or less MnS.

[0072] In some embodiments, the AI / ML management workflow includes a training phase, a simulation phase, a model deployment phase, and an inference phase. The following describes each of these four phases.

[0073] Training Phase: The purpose of the training phase is to train one or a set of AI / ML models. The training phase can include initial training and retraining. Optionally, the training phase also includes validation of the AI / ML entity to evaluate its performance on the training and validation data. If the validation results do not meet expectations (e.g., the variance is unacceptable), the AI / ML model associated with that entity needs to be retrained.

[0074] Simulation: During the simulation phase, AI / ML entities can be run in a simulation environment for inference. The purpose of the simulation phase is to evaluate the inference performance of AI / ML entities in the simulation environment before applying them to the target network or system, thus preventing AI / ML entities that do not meet the requirements from being deployed in the next phase.

[0075] Deployment phase: The process of making trained AI / ML entities available for use in target AI / ML inference functions. For example, an AI / ML model can be added to a function to enable AI / ML deployment.

[0076] Inference phase: The process of using AI / ML entities to perform reasoning through AI / ML inference functions. In other words, the inference phase is the stage where AI / ML actually works.

[0077] However, how to improve the simulation effect of AI / ML is an urgent problem to be solved.

[0078] It should be noted that the symbol " / " in this application can represent "and", "or", or and / or. For example, A / B can represent A and B, A or B, or A and / or B.

[0079] FIG3 is a schematic flow chart of a communication method 300 provided in an embodiment of the present application. Method 300 can improve the simulation effect of AI / ML. Method 300 is described below in conjunction with FIG3.

[0080] S310: A first device receives first information from a second device. Correspondingly, the second device sends the first information to the first device.

[0081] The first information may be used to instruct to simulate the capability of the AI / ML and / or to simulate the reasoning type of the AI / ML.

[0082] S320: The first device performs simulation on the AI / ML according to the first information to obtain a simulation report.

[0083] As an example, the first device may simulate the AI / ML capabilities and / or reasoning types as indicated by the first information, thereby obtaining a simulation report including simulation results of the AI / ML capabilities and / or reasoning types. As another example, the first device may train an ML model based on the AI / ML capabilities indicated by the first information, and run the ML model in a specific simulation environment to obtain a simulation report.

[0084] S330: The first device sends the simulation report to the second device. Correspondingly, the second device receives the simulation report from the first device.

[0085] Optionally, the first device may be a domain management function unit (Domain-MnF). The first device may also have other names, which are not limited in this application. For example, the first device may be called EMS, MAE, MnS producer, MDA producer, or have other names. For another example, the first device may be called a user, a requester, a vendor management function unit, an AI / ML MnS producer, etc.

[0086] Optionally, the second device may be a cross-domain management function unit (CD-MnF). The second device may also have other names, which are not limited in this application. For example, the second device may be called an NMS, an MnS consumer, an MDA consumer, or have other names. For another example, the second device may be called a provider, an operator management function unit, an AI / ML MnS consumer, etc.

[0087] Through the above embodiment, the first device can perform simulation according to the first information and send a simulation report, so that the device receiving the simulation report can make corresponding processing according to the simulation report, thereby improving the simulation effect. In addition, the first information can indicate the simulation of the capabilities and / or reasoning types of AI / ML. Among them, the capabilities of AI / ML can refer to the analysis capabilities of AI / ML for one or more indicators, and the reasoning type of AI / ML can refer to the ability of AI / ML to reason about a certain type of task. Through the first information, the simulation behavior of the first device can be indicated in a more fine-grained manner, thereby improving the simulation effect.

[0088] It should be noted that the term "AI / ML" in this application can be replaced by "AI," "ML," or "AIML," etc. AI / ML can be understood as AI and ML, as AI or ML, or as AI and / or ML.

[0089] In some optional embodiments, the first information may be used to indicate that the capabilities of the AI / ML are to be simulated. In other optional embodiments, the first information may be used to indicate that the reasoning type of the AI / ML is to be simulated. In still other optional embodiments, the first information may be used to indicate that both the capabilities and reasoning type of the AI / ML are to be simulated.

[0090] This application does not limit the message carried by the first information. For example, the first information can be carried in an ML simulation request (emulation request) message or other messages. This application does not limit the name of the first information. For example, the first information can be called a simulation indication, an inference simulation indication, an ML simulation indication, a simulation request, an inference simulation request, an ML simulation request, or has other names.

[0091] Exemplarily, simulating AI / ML can be understood as simulating the AI / ML model, or it can be understood as simulating the reasoning function deployed by the AI / ML model. For example, simulating the capabilities of AI / ML can be understood as simulating the AI / ML model, or it can be understood as simulating the ML model trained based on the capabilities of AI / ML, or it can be understood as simulating the capabilities of AI / ML for the AI / ML reasoning function, or it can be understood as simulating the ML model for the AI / ML reasoning function. For another example, simulating the reasoning type of AI / ML can be understood as simulating the AI / ML reasoning function to which the AI / ML model is deployed, or it can be understood as simulating the AI / ML reasoning function after the ML model trained based on the capabilities of AI / ML is deployed in the AI / ML reasoning function, or it can be understood as simulating the reasoning type of the AI / ML reasoning function.

[0092] This application does not limit the specific name of AI / ML capabilities. For example, AI / ML capabilities can also be called ML capability type, reasoning capability, AI capability, AI reasoning capability, ML capability, ML reasoning capability, or other names.

[0093] Optionally, in other implementation scenarios of the above embodiments, the AI / ML capabilities include at least one of the following: traffic analysis, coverage analysis, mobility analysis, load analysis, fault analysis, service experience analysis, energy efficiency analysis, or energy consumption analysis.

[0094] It should be noted that the AI / ML capabilities here can also be manufacturer-defined capabilities, and this application does not limit them here.

[0095] It should also be noted that the AI / ML capability here can be understood as the reasoning capability possessed by the AI / ML reasoning type, that is, the first device can train an ML model based on the AI / ML capability, and then apply it to the AI / ML reasoning type to complete the AI / ML reasoning function. Taking the AI / ML reasoning type as MRO in SON as an example, MRO, as the reasoning function of AI / ML, can use the mobility analysis ML model trained by the mobility analysis capability to perform reasoning, obtain the reasoning result, and then use the reasoning result to perform the MRO function. Other AI / ML reasoning types are similar to the above examples. You can replace the corresponding AI / ML reasoning type with the above examples, and I will not go into details here.

[0096] It is understood that traffic analysis capabilities can be embodied as an ML model with traffic analysis capabilities, or can be understood as being able to perform traffic analysis reasoning. That is, reasoning is performed using the ML model for traffic analysis, and the ML model for traffic analysis can analyze traffic indicators. The analysis of traffic indicators can include identifying traffic issues (e.g., traffic congestion), collecting statistics on traffic-related indicators, or predicting traffic-related indicators. Coverage analysis capabilities can be embodied as an ML model with coverage analysis capabilities, or can be understood as being able to perform coverage analysis reasoning. That is, reasoning is performed using the ML model for coverage analysis, and the ML model for coverage analysis can analyze coverage indicators. The analysis of coverage indicators can include identifying coverage issues (e.g., weak coverage, overcoverage, coverage holes, and out-of-area coverage), collecting statistics on coverage-related indicators, or predicting coverage-related indicators. Mobility analysis capabilities can be embodied as an ML model with mobility analysis capabilities, or can be understood as being able to perform mobility analysis reasoning. That is, reasoning is performed using the ML model for mobility analysis, and the ML model for mobility analysis can analyze mobility indicators. The analysis of mobility indicators may include the identification of mobility issues (e.g., premature handover, late handover, handover to the wrong cell, etc.), statistics on mobility-related indicators, or predictions of mobility-related indicators. The load analysis capability may be embodied as an ML model with load analysis capabilities, or it may be understood as being able to perform load analysis reasoning, i.e., using the load analysis ML model to perform reasoning, and the load ML model may analyze load indicators. The analysis of load indicators may include the identification of load issues (e.g., high load, low load), statistics on load-related indicators, or predictions of load-related indicators. The energy efficiency analysis capability may be embodied as an ML model with energy efficiency analysis capabilities, or it may be understood as being able to perform energy efficiency analysis reasoning, i.e., using the energy efficiency analysis ML model to perform reasoning, and the energy efficiency ML model may analyze energy efficiency indicators. The analysis of energy efficiency indicators may include statistics on energy efficiency-related indicators, or predictions of energy efficiency-related indicators. Energy consumption analysis can be embodied as an ML model with energy consumption analysis capabilities, or it can be understood as being able to perform reasoning on energy consumption analysis, that is, using the ML model for energy consumption analysis to perform reasoning, and the ML model for energy consumption can analyze energy consumption indicators. Among them, the analysis of energy consumption indicators can include statistics on energy consumption-related indicators or predictions of energy consumption-related indicators. Fault analysis capabilities can be embodied as an ML model with fault analysis capabilities, or it can be understood as being able to perform fault analysis and prediction, that is, using the ML model for fault analysis to perform reasoning, and the ML model for faults can analyze fault indicators. Among them, fault indicator analysis can include identification of fault problems (such as operational violations, physical violations, etc.), prediction of faults (such as faulty cells, etc.), statistics on fault-related indicators, etc.Service experience analysis capabilities can be reflected in ML models with service experience analysis capabilities. This can also be understood as the ability to perform statistical and predictive analysis of service experience. This means that the ML model used for service analysis performs inference, and the ML model for service experience can analyze relevant service experience indicators. This analysis of relevant service experience indicators can include identifying service experience issues (such as radio access network (RAN) and core network (CN) issues) and conducting statistical analysis of service experience indicators.

[0097] The first information is used to indicate the simulation of the AI / ML capability, which can be understood as the first information being used to indicate the analysis of at least one of the traffic, coverage, mobility, load, fault, service experience, energy efficiency or energy consumption of the AI / ML. It can also be understood as reasoning at least one of the ML model of traffic (analysis), the ML model of coverage (analysis), the ML model of mobility (analysis), the ML model of load (analysis), the ML model of fault (analysis), the ML model of service experience (analysis), the ML model of energy efficiency (analysis) or the ML model of energy consumption (analysis) of the AI / ML. Alternatively, it can also be understood as running (or using) at least one of the ML model of traffic (analysis), the ML model of coverage (analysis), the ML model of mobility (analysis), the ML model of load (analysis), the ML model of fault (analysis), the ML model of service experience (analysis), the ML model of energy efficiency (analysis) or the ML model of energy consumption (analysis) for reasoning in a simulation environment (or simulation stage).

[0098] The following describes some examples of specific indicators included in traffic, coverage, mobility, load, fault, service experience, energy efficiency, or energy consumption analysis. It should be noted that the parameters mentioned in this application may include the arithmetic mean, weighted average, maximum, or minimum value of the parameters. For example, the handover success rate may include the average handover success rate, the weighted average handover success rate, the maximum handover success rate, or the minimum handover success rate.

[0099] In some optional implementations, traffic analysis indicators may include at least one of: traffic congestion, burst traffic congestion, non-burst traffic congestion, cell / area average throughput, edge throughput, or traffic congestion recovery recommendations. This application does not limit the indicators included in traffic analysis; traffic indicators may also include other parameters.

[0100] Among them, network equipment (for example, base stations) can provide services for cells, and terminal equipment (for example, user equipment (UE)) can communicate with network equipment through the transmission resources used by the cell (for example, frequency domain resources, or spectrum resources). The cell can belong to a macro base station or a base station corresponding to a small cell. The small cells here can include: metro cells, micro cells, pico cells, femto cells, etc. Small cells are relative to macro cells. Macro cells generally have a larger coverage area (for example, a radius of more than 500 meters) and high transmission power, while small cells have a smaller coverage area (for example, a radius of tens of meters) and low transmission power, and are suitable for providing high-speed data transmission services. In addition, multiple cells can operate simultaneously on the same frequency on a carrier in an LTE system or a 5G system. In some special scenarios, the concepts of the above-mentioned carrier and cell can also be considered equivalent. For example, in a carrier aggregation (CA) scenario, when a secondary carrier is configured for a UE, the carrier index of the secondary carrier and the cell identification (cell ID) of the secondary cell operating on the secondary carrier are carried at the same time. In this case, the concepts of carrier and cell can be considered equivalent, for example, UE accessing a carrier is equivalent to accessing a cell.

[0101] A region may be the same as or different from a cell. For example, a region may be a RAN area. Another example is a region that includes multiple cells. Another example is a geographical area (e.g., an area determined by longitude and latitude) or an administrative area (e.g., a city).

[0102] In some optional implementations, the coverage analysis indicators may include: at least one of: weak coverage, overshooting, coverage holes, cross-region coverage, coverage problem area information, coverage problem cell identification, RSRP distribution, SINR distribution, RSRQ distribution, or coverage optimization suggestions. This application does not limit the indicators included in the coverage analysis, and the coverage indicators may also include other parameters.

[0103] Exemplarily, weak coverage may indicate that the signal at the boundary of a cell / area or other locations is weak. The reasons for weak coverage may be the coverage area required by the base station, the large distance between base stations, or obstruction by buildings. Exemplarily, over-coverage may indicate that the base station exceeds the set coverage range. Exemplarily, coverage holes may indicate that there are locations with weak signals in the cell / area. Exemplarily, out-of-zone coverage may indicate that the base station covers other cells (i.e., cells that do not belong to the base station). Exemplarily, RSRP may indicate the average value of the signal power received on all resource elements (REs) that carry a reference signal within a symbol. Exemplarily, SINR may indicate the ratio of the strength of the received useful signal to the strength of the received interference signal (noise and interference). Exemplarily, RSRQ may indicate the reception quality of the reference signal.

[0104] The RSRP distribution may also be referred to as RSRP signal strength distribution, signal strength distribution, or other names. The RSRQ distribution may also be referred to as signal quality distribution, RSRQ signal quality distribution, or other names.

[0105] In some optional implementations, the indicators of mobility analysis may include at least one of: handover success rate, handover failure rate, number of premature handovers, number of late handovers, number of handovers to incorrect cells, or mobility optimization recommendations. This application does not limit the indicators included in the mobility analysis, and mobility indicators may also include other parameters.

[0106] Exemplarily, the UE can switch from a source cell to a target cell. As a more specific example, the current cell of the UE is the source cell. After the target cell resources are successfully prepared, the base station of the target cell can notify the UE to switch to the target cell through an air interface reconfiguration message (such as a radio resource control (RRC) connection reconfiguration message). The UE can switch from the source cell to the target cell through the received air interface reconfiguration message. After the UE switches successfully, the base station of the target cell can notify the base station of the source cell to release the radio resources of the source cell and forward the unsent resources to the target cell. In addition, the target cell can also update the relationship between the user plane and the control plane nodes to serve the UE switched to the target cell.

[0107] Among them, the ratio between the number of successful UE handovers and the number of air interface reconfiguration messages sent by the base station can be used as the handover success rate. It should be noted that this application does not limit the specific calculation method of the handover success rate, and the handover success rate can also be reflected in other ways. For example, the handover success rate can be the ratio between the number of successful UE handovers and the number of handover requests sent by the original base station (the base station of the source cell) to the neighboring base station (the base station of the target cell).

[0108] The handover failure rate can be another aspect of the handover success rate. For example, the handover failure rate can be calculated by subtracting the handover success rate from 1. For another example, the ratio between the number of UE handover failures and the number of air interface reconfiguration messages sent by the base station can be used as the handover failure rate. It should be noted that this application does not limit the specific calculation method of the handover failure rate, and the handover failure rate can also be expressed in other ways.

[0109] The base station of the target cell can notify the UE of the handover time window. For example, the base station of the target cell can send an air interface resource reconfiguration message to the UE, which includes the handover time window (e.g., as reflected by the parameters of the t304 timer). In this way, if the UE switches before the handover time window, the handover is too early; if the UE switches after the handover time window, the handover is too late.

[0110] In some optional implementations, the load analysis indicators may include: cell PRB utilization, number of cell user connections, CPU usage, or at least one of load optimization recommendations. This application does not limit the indicators included in the load analysis, and the load indicators may also include other parameters.

[0111] The cell PRB utilization rate may also be referred to as the cell physical resource utilization rate or have other names. The number of cell user connections may also be referred to as the number of cell RRC connected users or have other names.

[0112] Exemplarily, the cell PRB utilization may include the cell uplink PRB utilization and / or the cell downlink PRB utilization.

[0113] In some optional implementations, the fault analysis indicators may include: at least one of the fault problem, fault cell identifier, fault severity, or fault recovery suggestion. This application does not limit the indicators included in the fault analysis, and the fault indicators may also include other parameters.

[0114] For example, the fault problem may include an operational violation problem and / or a physical violation problem. The fault problem may also include other contents.

[0115] In some optional implementations, the indicators for energy efficiency (or energy consumption) analysis may include at least one of: energy efficiency (or energy consumption) issues, identification of cells or base stations with energy efficiency (or energy consumption) issues, average throughput of cells / regions, available output power of base stations, number of users served at a specific power level, or energy-saving recommendations. This application does not limit the indicators included in the energy efficiency (or energy consumption) analysis, and energy efficiency (or energy consumption) indicators may also include other parameters.

[0116] Exemplarily, the energy efficiency issue may include high energy efficiency or low energy efficiency, and the energy consumption issue may include high energy consumption or low energy consumption. The available output power of the base station may also be referred to as the output power of the base station or have other names.

[0117] This application does not limit the specific name of the AI / ML inference type. For example, the AI / ML inference type may also be called ML inference type, AI inference type, ML inference simulation type, AI inference simulation type, AI / ML inference simulation type, or other names.

[0118] Through the above embodiments, the first information can instruct the first device to simulate at least one of the ML capabilities of traffic, coverage, mobility, load, fault, service experience, energy efficiency or energy consumption, thereby more finely indicating the simulation behavior of the first device and further improving the simulation effect, so as to evaluate the reasoning performance of the ML entity corresponding to the ML capability of traffic, coverage, mobility, load, fault, service experience, energy efficiency or energy consumption in the simulation environment before applying it to the destination network or system.

[0119] It should be noted that the capability of AI / ML may refer to the ability of AI / ML to analyze one or more indicators. This application does not limit the capabilities of AI / ML to the above-mentioned capabilities. The capabilities of AI / ML may also include the ability to analyze other types of indicators.

[0120] Optionally, in some other implementation scenarios of the above embodiment, the reasoning type of the AI / ML includes at least one of the following: the reasoning type of MDA, the reasoning type of SON, the reasoning type of NWDAF, or the reasoning type of radio access network intelligence (RAN intelligence).

[0121] The first information is used to indicate the simulation of the reasoning type of AI / ML, which can be understood as the first information being used to indicate the simulation of at least one of the reasoning types of MDA, SON, NWDAF, or RAN intelligence of AI / ML. It can be understood that the simulation of the reasoning type of AI / ML here is to run the ML model used for reasoning in the simulation environment. The ML model is the ML model corresponding to the AI / ML capability. It can also be understood that the ML model is the ML model obtained by training the AI / ML capability, so that the analysis or optimization performance of the ML model in the simulation environment can be evaluated before applying the ML model to the target reasoning type and further applying it to the existing network. The following introduces the reasoning types of MDA, SON, NWDAF, or RAN intelligence, as well as some examples of specific indicators of the reasoning type.

[0122] In some embodiments, the reasoning type of the MDA may include at least one of energy saving analysis, coverage problem analysis, fault analysis, network slice throughput analysis, slice load analysis, network slice traffic prediction analysis, service experience analysis, mobile performance analysis, or handover optimization analysis. The reasoning type of the MDA may also include other content. For specific indicators, please refer to the definition in Section 8.4 of TS28.104, which will not be repeated here. The following uses energy saving analysis and coverage analysis as examples to illustrate possible indicators.

[0123] Among them, the indicators of energy saving analysis may include at least one of the following: physical network function (PNF) power consumption (PNF power consumption), PNF energy consumption (PNF energy consumption), synchronization signal (SS)-RSRP distribution per SSB (beam) of serving NR cell, SS-RSRP distribution per SSB (beam) of neighbor NR cell, packet data convergence protocol (PDCP) data volume of NR cells, traffic load variation, UE throughput, delay related measurements of user plane function (UPF), data volume of UPF or virtual resource usage of NF.

[0124] The coverage analysis indicators may include at least one of the following: SS-RSRP distribution of each SSB (or SSB beam) of the serving NR cell, SS-RSRP distribution of each SSB (or SSB beam) of the neighboring NR cell, wideband channel quality indicator (CQI) distribution, RSRP distribution per neighboring E-UTRAN cell, power headroom distribution for NR cell, wideband CQI distribution for NR cell, timing advance distribution for NR cell, number of UE context release request (gNB-DU initiated), number of UE context release request per SSB (gNB-DU initiated), and number of UE context release request per SSB (gNB-DU initiated). initiated), number of UE context release requests (gNB-CU initiated), number of UE context release requests per SSB (gNB-CU initiated), next generation,RSRP related measurements for ng-eNB, UE power headroom related measurements for ng-eNB, wideband CQI distribution for ng-eNB, average sub-band CQI for ng-eNB, UE Rx-Tx time difference related measurements for ng-eNB, angle of arrival (AOA) related measurements for ng-eNB, timing advance distribution for ng-eNB, or number of UE context release request initiated by ng-eNodeB.

[0125] This application does not limit the indicators corresponding to the reasoning type of MDA, and the reasoning type of MDA may also correspond to other indicators.

[0126] In some embodiments, the reasoning type of SON may include at least one of mobility robustness optimization (MRO), distributed mobility robustness optimization (DMRO), energy saving (ES), distributed energy saving (DES), mobility load balancing (MLB), or distributed mobility load balancing (DMLB).

[0127] In some embodiments, the reasoning type of RAN intelligence may include at least one of MRO, DMRO, ES, DES, MLB, DMLB, mobility optimization (MO), network energy saving (NES), and load balancing (LB). It should be noted that in related technical solutions, SON use cases may not have AI / ML capabilities, but in the solution of this application, SON use cases can have AI / ML capabilities, that is, they can implement the solution of this application.

[0128] It should be noted that this application does not limit the names of the above parameters, and the above parameters can also have other names.

[0129] In some embodiments, the performance evaluation indicators of the ES may include at least one of the following: PNF energy consumption, downlink or uplink data volume of the PDCP SDU carried by the data radio resource (DRB.PdcpSduVolumeDL_Filter, DRB.PdcpSduVolumeUL_Filter), downlink data volume of the cell PDCP SDU on the X2 interface (DL Cell PDCP SDU Data Volume on X2 Interface), downlink data volume of the cell PDCP SDU on the Xn interface (DL Cell PDCP SDU Data Volume on Xn Interface), uplink data volume of the cell PDCP SDU on the X2 interface (UL Cell PDCP SDU Data Volume on X2 Interface), uplink data volume of the UL cell PDCP SDU on the Xn interface (UL Cell PDCP SDU Data Volume on Xn Interface), and PDCP on the F1 interface carried by the data radio resource. The downlink or uplink data volume of the SDU (DRB.F1uPdcpSduVolumeDL_Filter or DRB.F1uPdcpSduVolumeUL_Filter), the downlink or uplink data volume of the PDCP SDU of the Xn interface carried by the data radio resource (DRB.XnuPdcpSduVolumeDL_Filter or DRB.XnuPdcpSduVolumeUL_Filter), or the downlink or uplink data volume of the PDCP SDU of the X2 interface carried by the data radio resource (DRB.X2uPdcpSduVolumeDL_Filter,, or DRB.X2uPdcpSduVolumeUL_Filter).

[0130] Where DRB stands for data radio bearer (DRB), SDU stands for service data unit (SDU), DL stands for downlink (DL), and UL stands for uplink (UL). For the specific definitions of the above indicators, please refer to the relevant content in Table 6.2.2.2.3.2-1 of 3GPP TS28.310.

[0131] In some embodiments, the performance evaluation index of the MRO may include at least one of the following: the number of handover events, the number of handover failures, the radio access technology,number of intra-RAT handover events, number of intra-RAT handover failures, number of inter-RAT handover events, number of inter-RAT handover failures, number of intra-RAT too late handover failures, number of intra-RAT too early handover failures, number of intra-RAT handover failures to wrong cell, number of intra-RAT too late handover failures per source beam, number of intra-RAT too early handover failures per source beam, number of intra-RAT handover failures to wrong cell per source beam beam), number of inter-RAT too late handover failures, number of inter-RAT too early handover failures, number of unnecessary handover to another RAT, or number of inter-RAT handover ping pong.

[0132] In some embodiments, the performance evaluation indicators of MRO may include at least one of the following: downlink total PRB usage (DL total PRB usage), uplink total PRB usage (UL total PRB usage), downlink total PRB usage distribution (distribution of DL total PRB usage), uplink total PRB usage distribution (distribution of UL total PRB usage), downlink PRB used for data traffic (DL PRB used for data traffic), uplink PRB used for data traffic (UL PRB used for data traffic), average number of RRC connections (mean number of RRC connections), maximum number of RRC connections (max number of RRC connections), average number of stored inactive RRC connections (mean number of stored inactive RRC connections), or maximum number of stored inactive RRC connections (max number of stored inactive RRC connections).

[0133] This application does not limit the indicators corresponding to the reasoning types of SON or RAN intelligence. SON or RAN intelligence may also correspond to other indicators.

[0134] In some embodiments, the performance evaluation metrics of the reasoning type of NWDAF may include: the number of subscriptions and / or requests, the number of notifications and / or responses, or the time consumption of NWDAF generating analytics result.

[0135] This application does not limit the indicators corresponding to the reasoning types of NWDAF, and the reasoning types of NWDAF may also correspond to other indicators. The reasoning types of NWDAF can be found in the description of the relevant technical solutions and will not be repeated here.

[0136] For example, when the AI / ML reasoning type is MRO, the first information is used to indicate the simulation of the MRO. This means deploying the ML model corresponding to the AI / ML capabilities that the MRO can use into the MRO function, and running the MRO function in the simulation environment to obtain the MRO optimization results and further obtain the MRO performance indicators. Other reasoning types are similar to the above examples. Simply replace the corresponding reasoning type with the above examples, and we will not elaborate on them here.

[0137] Through the above embodiments, the first information can instruct the first device to analyze at least one reasoning type among the management data analysis function, the self-optimizing network function, the network data analysis function, or the wireless access network intelligent function, thereby more finely indicating the simulation behavior of the first device, thereby further improving the simulation effect, so as to evaluate the performance of the management data analysis function, the self-optimizing network function, the network data analysis function, or the wireless access network intelligent function in the simulation environment before applying it to the target network or system.

[0138] It should be noted that the reasoning type of AI / ML may refer to the ability of AI / ML to reason about a certain type of task. This application does not limit the reasoning type of AI / ML to only the above content. The reasoning type of AI / ML may also include the ability to reason about other types and their indicators.

[0139] Optionally, in some other implementation scenarios of the above embodiments, the first information also includes at least one of the following: a model identifier of the AI / ML; a model identifier group of the AI / ML; an activation status, used to indicate whether to turn on or off the simulation of the AI / ML; a simulation area, used to indicate the area where the AI / ML is simulated; a simulation time, used to indicate the time when the AI / ML is simulated; a first request, used to request the simulation result of the capability of the AI / ML; a second request, used to request the simulation result of the reasoning type; cycle information, used to indicate the sending cycle of the simulation report; the performance indicator of the capability of the AI / ML; the performance indicator of the reasoning type of the AI / ML model; a simulation strategy, including at least one of the performance monitoring threshold, performance monitoring area or performance monitoring time of the AI / ML model; or a termination strategy, used to indicate the conditions for terminating the simulation of the AI / ML.

[0140] It should be noted that, in some other optional embodiments, the first information in method 300 may include at least one of the following: a first indication for indicating that the capability of the AI / ML is to be simulated; a second indication for indicating that the reasoning type of the AI / ML is to be simulated; a model identifier of the AI / ML; a model identifier group of the AI / ML; an activation state for indicating whether the simulation of the AI / ML is to be turned on or off; a simulation area for indicating the area in which the AI / ML is to be simulated; a simulation time for indicating the time when the AI / ML is to be simulated; a first request for requesting the simulation result of the capability of the AI / ML; a second request for requesting the simulation result of the reasoning type; period information for indicating the sending period of the simulation report; a performance indicator of the capability of the AI / ML; a performance indicator of the reasoning type of the AI / ML model; a simulation strategy including at least one of a performance monitoring threshold, a performance monitoring area, or a performance monitoring time of the AI / ML model; or a termination strategy for indicating the conditions for terminating the simulation of the AI / ML. In other words, the first indication and the second indication are not necessarily included in the first information in method 300.

[0141] In addition, it should be noted that any content in the first information may not be carried in the first information, and may be sent independently or carried in other information.

[0142] The AI / ML model identifier can indicate which model is used to perform the inference simulation. The AI / ML model identifier group can indicate which group of models is used to perform the inference simulation. For example, the AI / ML model identifier group can be a list of model identifiers or a model group identifier.

[0143] The model can be an ML model or an AI model. The model can also be represented by an entity, such as an ML entity. The model group can be generated after joint training or joint testing.

[0144] As an example, mLEntityToEmulationRef can be used to represent an instance identifier of an ML entity, which can be a model that has been trained in a training function (e.g., the first device or another device). The instance identifier can be, for example, a distinguished name (DN). The first device can perform simulation on the ML entity based on the first information.

[0145] As an example, mLEntityCoordinationGroupToEmulationRef can be used to represent the instance identifiers of a group of ML entities (or a group of instance identifiers of multiple ML entities), where the group of ML entities can include at least one ML entity, and the ML entity in the at least one ML entity can be a model that has been trained in a training function (e.g., the first device or another device). The instance identifiers of a group of ML entities can be a list of instance identifiers, such as a list of DNs, or an instance group identifier.

[0146] The activation status can be used to indicate whether the simulation of the AI / ML is turned on or off. For example, the activation status can indicate activated, indicating that the simulation of the AI / ML is turned on; the activation status can indicate deactivated, indicating that the simulation of the AI / ML is turned off. For another example, the activation status can indicate true, indicating that the simulation of the AI / ML is turned on; the activation status can indicate false, indicating that the simulation of the AI / ML is turned off.

[0147] This application does not limit the name of the activation state, and the activation state may also be called an inference simulation switch or have other names.

[0148] It should be noted that turning on the simulation of AI / ML does not mean that the first device will definitely send a simulation report. The first device may also only simulate the AI / ML without sending a simulation report. Turning off the simulation of AI / ML does not mean that the first device will definitely not send a simulation report. For example, the first device may generate a simulation report based on the previous simulation of AI / ML and send the simulation report. For another example, the first device may send a simulation report, and the simulation report indicates that the current simulation status is that simulation is not turned on.

[0149] The emulation area can be used to indicate the area where the AI / ML is simulated. For example, the emulation area can be a base station identifier or a list of base station identifiers. For another example, the emulation area can be a cell identifier or a list of cell identifiers. For another example, the emulation area can be longitude and latitude. It is understandable that longitude and latitude can represent a geographical area. This application does not limit the specific form of the emulation area, and the emulation area can also have other identifiers. For example, the identifier of an administrative area, etc.

[0150] It should be noted that for scenarios where AI / ML is deployed on base stations, the simulation area can indicate a list of cells / base stations that are allowed to participate in the simulation.

[0151] This application does not limit the name of the simulation area, and the simulation area may also be called an inference simulation area or have other names.

[0152] The simulation time can be used to indicate the duration of the AI / ML simulation. The simulation time can be expressed as a time period, duration, or time cycle. For example, the simulation time can be a period of time with a start and end time, indicating that the first device must perform simulation within that time period. Another example can be a numerical value representing a time length, indicating that after the first device starts simulation, it must complete the simulation within that time period. Another example can be a numerical value representing a time cycle, indicating that the first device must perform simulation once every time period.

[0153] This application does not limit the name of simulation time, and simulation time can also be called inference simulation time or have other names.

[0154] The first request may be used to request simulation results of the AI / ML capability.

[0155] Exemplarily, the first request may request simulation results of one or more capability types of the AI / ML. For example, the first request may request the first device to simulate at least one of traffic analysis, coverage analysis, mobility analysis, load analysis, fault analysis, service experience analysis, energy efficiency analysis, or energy consumption analysis of the AI / ML, and return the corresponding simulation results.

[0156] This application does not limit the name of the first request; the first request may also be referred to as a capability result request or have other names. In some optional embodiments, the first request may be carried as a field in the request message. For example, the first request may be carried in the request message as an emulation ML capability result. In this way, the request message can be used to request the simulation result of the AI / ML capability.

[0157] The second request may be used to request a simulation result of the inference type.

[0158] As an example, the second request may request simulation results of one or more reasoning types of AI / ML. For example, the second request may request the first device to simulate AI / ML based on at least one management use case of MDA, SON, NWDAF, or radio access network intelligence, and return a corresponding simulation report. For another example, the second request may request the first device to simulate AI / ML based on at least one management use case of MRO, DMRO, ES, DES, MLB, DMLB, MLB, LB, and NES, and return a corresponding simulation report.

[0159] As another example, the second request may request a simulation result of at least one indicator of one or more reasoning types of AI / ML. Examples of indicators of reasoning types can be found above and are not repeated here.

[0160] This application does not limit the name of the second request; the second request may also be referred to as an inference type result request or have other names. In some optional embodiments, the second request may be carried as a field in the request message. For example, the second request may be carried in the request message as an inference emulation type result. In this way, the request message can be used to request a simulation result of this inference type.

[0161] It should be noted that the first request and the second request can be carried in one request message. As an example, the first request and the second request can be a field of the request message. For example, the request message can carry an inference simulation result field, which can request the simulation result of the AI / ML capability and the simulation result of the inference type. As another example, the first request and the second request can be two fields of the request message respectively. For example, the first request is carried in the request message as the simulation ML capability result, and the second request is carried in the request message as the inference simulation type result.

[0162] However, the present application does not limit the carrying format of the first request and the second request. For example, the first request and the second request may also be carried in different request messages respectively.

[0163] The period information may be used to indicate the sending period of the simulation report. In other words, the period information may indicate the reporting period of the simulation result.

[0164] It should be noted that, unless otherwise specified, this application does not distinguish between performance indicators and indicators. In other words, performance indicators and indicators can be used interchangeably.

[0165] Exemplarily, the performance indicators of AI / ML capabilities may include: traffic congestion problems, burst traffic congestion, non-burst traffic congestion, average cell / area throughput, edge throughput, traffic congestion recovery suggestions, weak coverage, over-coverage, coverage holes, cross-area coverage, coverage problem area information, coverage problem cell identification, RSRP distribution, SINR distribution, RSRQ distribution, coverage optimization suggestions, switching success rate, switching failure rate, number of early switches, number of late switches, number of switches to wrong cells, mobility optimization suggestions, cell PRB utilization, number of cell user connections, CPU usage, load optimization suggestions, fault problems, fault problem cell identification, fault severity, fault recovery suggestions, energy efficiency problems, energy efficiency problem cell or base station identification, energy consumption problems, energy consumption problem cell or base station identification, available output power of the base station, or the number of users served at a specific power, energy saving suggestions.

[0166] For example, the performance indicator of the inference type of the AI / ML model may include at least one of the following:

[0167] At least one indicator for different use cases in MDA;

[0168] At least one metric for different use cases in SON;

[0169] At least one indicator for different use cases in NWDAF;

[0170] At least one metric for different use cases in radio access network intelligence.

[0171] MDA use cases may include energy saving analysis, coverage problem analysis, fault analysis, network slice throughput analysis, slice load analysis, network slice traffic prediction analysis, service experience analysis, mobile performance analysis, or handover optimization analysis. SON use cases or wireless access network intelligence use cases may include at least one of MRO, DMRO, ES, DES, MLB, DMLB, LB, MLB, and NES. The indicators for the above reasoning types are described above and are not detailed here.

[0172] The simulation strategy may include at least one of performance monitor thresholds, performance monitor area, or performance monitor time of the AI / ML model.

[0173] A performance monitoring threshold can be a numerical value or a numerical range. A performance indicator satisfies the performance monitoring threshold if the magnitude relationship between the performance indicator and the performance monitoring threshold satisfies a preset magnitude relationship. For example, if the preset magnitude relationship is that the performance indicator is greater than or equal to the performance monitoring threshold, then the performance indicator satisfies the performance monitoring threshold if the performance indicator is greater than or equal to the performance monitoring threshold.

[0174] The description of the performance monitoring area can be found in the description of the simulation area above, and will not be repeated here. The description of the performance monitoring time can be found in the description of the simulation time above, and will not be repeated here.

[0175] In some optional embodiments, S320 may include: the first device simulating the AI / ML according to the simulation strategy. In other words, based on the first information indicating that the AI / ML capability and / or reasoning type is to be simulated, the first device further determines whether to simulate the AI / ML according to the simulation strategy.

[0176] As an example, the simulation strategy includes a performance monitoring threshold, and when the performance indicator meets the performance monitoring threshold, the first device performs simulation on AI / ML.

[0177] As another example, the simulation strategy includes a performance monitoring threshold and a performance monitoring area, and when a performance indicator in the performance monitoring area meets the performance monitoring threshold, the first device performs simulation on AI / ML.

[0178] As another example, the simulation strategy includes a performance monitoring threshold and a performance monitoring time. When the performance indicator within the performance monitoring time meets the performance monitoring threshold, the first device performs simulation on AI / ML.

[0179] As another example, the simulation strategy includes a performance monitoring threshold, a performance monitoring area, and a performance monitoring time. When the performance indicator within the performance monitoring time and within the performance monitoring area meets the performance monitoring threshold, the first device performs simulation on AI / ML.

[0180] As another example, the simulation strategy includes a performance monitoring area and / or a performance monitoring time, and the first device performs simulation on AI / ML during the performance monitoring time and within the performance monitoring area.

[0181] In other optional embodiments, S320 may be replaced with: the first device simulating the AI / ML according to the simulation policy. In other words, regardless of whether the first device receives an instruction to simulate the capabilities and / or reasoning types of the AI / ML, the first device may independently determine whether to simulate the AI / ML according to the simulation policy.

[0182] The example of the first device performing simulation on AI / ML according to the simulation strategy can be found above and will not be repeated here. Moreover, based on the replacement of S320 with the above solution, the new solution can be combined with other embodiments of the present application and can be applied to other descriptions of the present application.

[0183] This application does not limit the name of the simulation policy. The simulation policy may also be called an ML simulation policy, a monitoring policy, or have other names.

[0184] The termination policy may be used to indicate a condition for terminating the simulation of the AI / ML. In some optional embodiments, the method 300 further includes: the first device terminating the simulation of the AI / ML according to the termination policy.

[0185] For example, the termination policy may include a second device (e.g., CD-MnF) requesting termination. Another example may include a reasoning function stop, meaning that when the reasoning function in the first device is stopped, the first device may terminate the AI / ML simulation. Another example may include an AI / ML model update, meaning that when the ML model in the reasoning function in the first device is being updated, the first device may terminate the AI / ML simulation.

[0186] This application does not limit the name of the termination policy. The termination policy may also be called an emulation termination condition, an emulation termination policy, or have other names.

[0187] In some optional embodiments, the first information may further include a process monitoring instruction. The process monitoring instruction may be used to indicate monitoring or control of the AI / ML simulation process. For example, the process monitoring instruction may include a cancel process, a suspend process, and a resume process. The cancel process may instruct the first device to cancel the simulation process; the suspend process may instruct the first device to suspend the simulation process; and the restart process may instruct the first device to restart the simulation process.

[0188] Through the above embodiment, the first information contains more information, thereby more precisely indicating the simulation behavior of the first device, thereby further improving the simulation effect.

[0189] Optionally, in some other implementation scenarios of the above embodiments, the simulation report includes at least one of the following: a model identification of the AI / ML; a model identification group of the AI / ML; a simulation status, used to indicate the status of the simulation of the AI / ML; a simulation reasoning capability result, used to indicate the simulation result of the capability of the AI / ML; or a simulation reasoning type result, used to indicate the simulation result of the reasoning type of the AI / ML.

[0190] The meanings of the AI / ML model identifier and the AI / ML model identifier group can be found in the previous description and will not be repeated here.

[0191] The emulation status can be used to indicate the status of the simulation of the AI / ML, including the status of the simulation of the capability of the AI / ML, and / or the simulation status of the simulation type of the AI / ML. For example, the simulation status may include "executed", indicating that the first device has completed the simulation of the AI / ML. For another example, the simulation status may include "not executed", indicating that the first device has not performed simulation on the AI / ML. For another example, the simulation status may include "executing", indicating that the first device is performing simulation on the AI / ML. For another example, the simulation status may include "executing", indicating that the first device is performing simulation on the AI / ML. For another example, the simulation status may include "execution", indicating that the first device is performing simulation on the AI / ML. For another example, the simulation status may include "execution percentage", indicating the progress of the first device performing simulation on the AI / ML. Among them, when the execution percentage is 0, it means not executed; when the execution percentage is 1, it means executed; when the execution percentage is between 0 and 1, it means executing. It should be noted that the simulation status can be the simulation status of the capability of the AI / ML, or the simulation status of the reasoning type of the AI / ML.

[0192] The simulation reasoning capability result can be used to indicate the simulation result of the capability of the AI / ML.

[0193] In some optional embodiments, the simulation reasoning capability results may include at least one of traffic analysis, coverage analysis, mobility analysis, load analysis, fault analysis, service experience analysis, energy efficiency analysis, or energy consumption analysis.

[0194] For each type of capability, the simulation reasoning capability result may include some or all indicators of the capability of that type.

[0195] Optionally, in some other implementation scenarios of the above embodiments, the simulation reasoning capability results include at least one of the following: traffic congestion problems, burst traffic congestion, non-burst traffic congestion, cell / area average throughput, edge throughput, traffic congestion recovery suggestions, weak coverage, over-coverage, coverage holes, cross-area coverage, coverage problem area information, coverage problem cell identification, RSRP distribution, SINR distribution, RSRQ distribution, coverage optimization suggestions, switching success rate, switching failure rate, number of early switches, number of late switches, number of switches to wrong cells, mobility optimization suggestions, cell PRB utilization, number of cell user connections, CPU usage, load optimization suggestions, fault problems, fault problem cell identification, fault severity, fault recovery suggestions, energy efficiency problems, energy efficiency problem cell or base station identification, energy consumption problems, energy consumption problem cell or base station identification, available output power of the base station, or the number of users served at a specific power, energy-saving suggestions.

[0196] The specific meanings of the above indicators can be found in the previous article and will not be repeated here.

[0197] Through the above embodiments, the simulation report may include one or more performance indicators of traffic analysis, coverage analysis, mobility analysis, load analysis, fault analysis, service experience analysis, energy efficiency analysis or energy consumption analysis, so that the device receiving the simulation report can make corresponding processing according to the various indicators of AI / ML capabilities, thereby further improving the simulation effect.

[0198] In addition, it should be noted that any content in the simulation report may not be carried in the simulation report, but may be sent independently or carried in other information.

[0199] In some optional embodiments, the simulation reasoning capability results may also include the performance results of the model used to reason about the AI / ML capability. For example, the above performance results may include at least one of accuracy, precision, or recall. Among them, the accuracy of the model may refer to the percentage of correct results inferred to the total samples. The precision of the model may refer to the ratio of the correct results inferred to the actual correct samples. The recall rate of the model may refer to the ratio of the number of positive samples inferred in the reasoning result to the number of true positive samples. Among them, the positive sample may refer to the target category corresponding to the true value. The above reasoning results may also be referred to as reasoning simulation results.

[0200] The above performance results may further include an F1 score, which may be a harmonic mean of precision and recall.

[0201] The simulation reasoning type result can be used to indicate the simulation result of the reasoning type of the AI / ML.

[0202] In some optional embodiments, the simulation reasoning type result may include at least one of the use cases of the MDA function, the SON function, the NWDAF function, or the wireless access network intelligent function. The use cases of the MDA function may include energy saving analysis, coverage problem analysis, fault analysis, network slice throughput analysis, slice load analysis, network slice traffic prediction analysis, service experience analysis, mobile performance analysis, handover optimization analysis, etc. The use cases of the SON function or the wireless access network intelligent function may include at least one of MRO, DMRO, ES, DES, MLB, DMLB, MLB, LB, and NES.

[0203] Optionally, in other implementation scenarios of the above embodiments, the simulation results of the AI / ML reasoning type include at least one of the following: the identification of the cell / base station, information of the shut-down cell / carrier / time slot, CIO, TTT, or performance indicators of the reasoning type.

[0204] For example, the simulation results of DMRO, MO, or MRO may include at least one of the cell / base station identifier, CIO, or TTT. For another example, the simulation results of DES, NES, and ES may include the cell / base station identifier and / or information about shutting down the cell / carrier / time slot. For another example, the simulation results of DMLB, LB, and MLB may include the cell / base station identifier and / or CIO.

[0205] In some optional embodiments, the simulation reasoning capability result may further include the value of a performance indicator corresponding to at least one reasoning type. The indicators of the above reasoning types are described above and are not described here in detail.

[0206] In some optional embodiments, the simulation result may also include a progress status, which may indicate the state of the first device simulation process. In other optional embodiments, the progress status may not be included in the simulation result, but may be sent separately or included in other information.

[0207] Through the above embodiments, the simulation report may include at least one simulation result and / or performance indicator of the reasoning type, so that the device receiving the simulation report can make corresponding processing according to the simulation result or various indicators of the AI / ML reasoning type, thereby further improving the simulation effect.

[0208] Furthermore, through the above embodiments, the simulation report contains more information, so that the device receiving the simulation report can make corresponding processing, thereby further improving the simulation effect.

[0209] In some optional embodiments, method 300 further includes: the first device sending reasoning simulation capabilities to the second device. The reasoning simulation capabilities may include the AI / ML reasoning types supported by the first device and / or the AI / ML capabilities supported by the first device.

[0210] It should be noted that the first device may proactively send the reasoning simulation capability to the second device, or may send the reasoning simulation capability to the second device based on a request from the second device. For example, in some optional embodiments, method 300 further includes: the first device receiving a capability discovery request from the second device. The capability discovery request is used to request the reasoning simulation capability of the first device.

[0211] Figure 4 is a schematic diagram of an implementation model of AI / ML simulation provided by the present application. The implementation model shown in Figure 4 can be combined with the above-mentioned method 300. In Figure 4, the solid diamond can represent an inclusion relationship, for example, the simulation function object contains the ML simulation request object, etc. The arrow can represent an association relationship, for example, the association between the ML simulation request object, the ML simulation process object and the ML entity object. The cardinality relationship between the intent object attributes with inclusion and association relationships can be expressed as: 1, associated with 1 object attribute; *, associated with 0 or more object attributes; 1...*, associated with at least one object attribute. For more specific definitions of inclusion and association relationships, please refer to 3GPP TS 32.156, which will not be repeated in this application. The implementation model is introduced below in conjunction with Figure 4.

[0212] The implementation model includes multiple instance (or information) object classes (IOC) related to simulation. The implementation model includes an object ML simulation function (MLEmulationFunction) < <ioc>>、ML simulation process (MLEmulationProcess)< <ioc>>、ML simulation request (MLEmulationRequest)< <ioc>>、ML simulation report (MLEmulationReport)< <ioc>> and ML simulation policy (MLEmulationPolicy)< <ioc>>. The above names may be other names, which are not limited in this application.

[0213] Among them, IOC MLEmulationRequest, IOC MLEmulationReport, IOC MLEmulationProcess and IOC MLEmulationPolicy are included in the simulation function (EmulationFunction) object. EmulationFunction can be a proxy class (proxyclass). Exemplarily, EmulationFunction can represent one or more of the following: object ML simulation function (MLEmulationFunction) or object AI ML inference capability (AIMLInferenceCapability). Among them, the object MLEmulationFunction can be used to represent the simulation function, that is, the bearer of the simulation function. AIMLInferenceCapability can be used to represent the AI / ML inference function (AI / ML inference function), and can also be understood as the bearer of the inference function.

[0214] Among them, the IOC MLEmulationFunction can be contained in the ML simulation entity (MLEmulationEntity) object. Among them, MLEmulationEntity can be a proxy class. Exemplarily, MLEmulationEntity can represent one or more of the following: a group of managed entities, a group of managed functions (Managed Function) or a group of managed management functions (Management Function). Among them, a group of managed entities can be represented by a network area, such as a subnet (SubNetwork); a group of managed entities can include a group of base stations. A group of managed functions can be a communication function implemented by software running on dedicated hardware, or can represent a communication function implemented by software running on a network function virtualization infrastructure (NFVI). A group of managed functions can also be a virtual network function of a base station or a core network. A group of managed management functions can be an inference function (AI / ML inference function), which can also be understood as the bearer of the inference function.

[0215] IOC MLEmulationRequest, IOC MLEmulationReport, IOC MLEmulationProcess, and IOC MLEmulationPolicy may be associated with MLEntity objects and MLEntityCoordinationGroup objects.

[0216] The following describes the parameters or attributes of each object in conjunction with Figure 4 and some tables. It should be noted that the parameters or attributes described below can be one or more, and this application does not limit this.

[0217] For example, MLEmulationFunciton< <ioc>The properties of > are shown in Table 1. The IOC represents the logical function that can be used to perform simulation on an ML entity or a group of ML entities.

[0218] Table 1

[0219] The description of activationStatus can refer to the description of activation status in the aforementioned method 300, which is not repeated here.

[0220] For example, MLEmulationRequest< <ioc>The properties of > are shown in Table 2.

[0221] Table 2

[0222] The description of the above parameters can refer to the corresponding description in the above method 300, which will not be repeated here. <ioc>> can include all or part of the properties in Table 2. In other words, MLEmulationRequest< <ioc>>The attributes can include at least one row of any one of Table 2.

[0223] For example, MLEmulationReport< <ioc>The properties of > are shown in Table 3.

[0224] Table 3

[0225] The description of the above parameters can refer to the corresponding description in the above method 300, which will not be repeated here. <ioc>> can include all or part of the properties in Table 3. In other words, MLEmulationReport< <ioc>>'s attributes may include at least one row from any of Table 3.

[0226] For example, MLEmulationPolicy< <ioc>The properties of > are shown in Table 4.

[0227] Table 4

[0228] The description of the above parameters can refer to the corresponding description in the above method 300, which will not be repeated here. <ioc>> can include all or part of the attributes in Table 4. In other words, MLEmulationPolicy< <ioc>>'s attributes may include at least one row from any of Table 4.

[0229] For example, MLEmulationProcess< <ioc>The properties of > are shown in Table 5.

[0230] Table 5

[0231] The description of the above parameters can refer to the corresponding description of the simulation results and the first information in the above method 300, which will not be repeated here. <ioc>> can include all or part of the properties in Table 5. In other words, MLEmulationProcess< <ioc>>'s attributes may include at least one row from any of Table 5.

[0232] The following is an introduction to the device embodiment corresponding to the method embodiment of the present application. The following is only a brief introduction to the device, and the specific implementation steps and details of the solution can be referred to the method embodiment above.

[0233] To implement the various functions of the method provided herein, both the first device and the second device may include hardware structures and / or software modules, and implement the aforementioned functions in the form of hardware structures, software modules, or a combination of hardware structures and software modules. Whether a particular one of the aforementioned functions is implemented in the form of hardware structures, software modules, or a combination of hardware structures and software modules depends on the specific application and design constraints of the technical solution.

[0234] 5 is a schematic block diagram of a communication device 500 according to an embodiment of the present application. The communication device 500 includes a processor 510 and a communication interface 520, which may be interconnected via a bus 530. The communication device 500 may be a first device or a second device.

[0235] Optionally, the communication device 500 may further include a memory 540. The memory 540 includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or portable read-only memory (CD-ROM), and is used for related instructions and data.

[0236] The processor 510 may be one or more CPUs. If the processor 510 is a CPU, the CPU may be a single-core CPU or a multi-core CPU. The processor 510 may be a signal processor, a chip, or other integrated circuit that can implement the method of the present application, or a portion of the circuitry used for processing functions in the aforementioned processor, chip, or integrated circuit. In addition, the communication interface 520 may also be referred to as an input / output interface or a communication interface. The communication interface 520 is used for input or output of signals or data, and may also be an input / output circuit.

[0237] When the communication device 500 is the first device, the communication device 500 illustratively includes a processor 510 and a communication interface 520. The communication interface 520 is configured to receive first information; the processor 510 is configured to simulate the AI / ML based on the first information to obtain a simulation report; and the communication interface 520 is further configured to send the simulation report.

[0238] When the communication device 500 is the second device, illustratively, the communication device 500 includes a communication interface 520. The communication interface 520 is used to send the first information and receive the simulation report.

[0239] When the processor 510 is the first device, illustratively, the processor 510 is used to receive first information from the communication interface 520; to perform simulation on the AI / ML according to the first information to obtain a simulation report; and to send the simulation report to the communication interface 520.

[0240] In other words, the processor 510 may receive the first information through the communication interface 520 , and send the simulation report through the communication interface 520 .

[0241] When the processor 510 is the second device, illustratively, the processor 510 is configured to send the first information to the communication interface 520 ; and to receive the simulation report from the communication interface 520 .

[0242] In other words, the processor 510 may send the first information through the communication interface 520 , and receive the simulation report through the communication interface 520 .

[0243] The above description is merely exemplary. For details, please refer to the contents of the above method embodiments. The implementation of each operation in FIG5 may also correspond to the corresponding description of the method embodiment shown in FIG3 or FIG4.

[0244] Figure 6 is a schematic block diagram of another communication device 600 according to an embodiment of the present application. Communication device 600 can be the first device or the second device, or a chip or module within the first device or the second device, configured to implement the methods described in the above embodiments. Communication device 600 includes a transceiver unit 610. The following provides an exemplary description of transceiver unit 610.

[0245] The transceiver unit 610 may include a transmitting unit and a receiving unit. The transmitting unit is configured to execute a transmitting operation of the communication device, and the receiving unit is configured to execute a receiving operation of the communication device. For ease of description, this embodiment of the application combines the transmitting unit and the receiving unit into a single transceiver unit. This is described here as a unified description and will not be repeated later.

[0246] When the communication device 600 is a first device, illustratively, the transceiver unit 610 is configured to receive first information.

[0247] Optionally, the communication device 600 may further include a processing unit 620, which is used to execute the content of the first device involving processing, coordination and other steps.

[0248] When the communication device 600 is the second device, illustratively, the transceiver unit 610 is configured to send the first information.

[0249] Optionally, the communication device 600 may further include a processing unit 620, which is used to execute the content of the second device involving processing, coordination and other steps.

[0250] The above contents are merely exemplary descriptions. When the communication device 600 is the first device or the second device, it will be responsible for executing the methods or steps related to the first device or the second device in the above method embodiments.

[0251] Optionally, the communication device 600 further includes a storage unit 630, which is used to store a program or code for executing the aforementioned method.

[0252] The device embodiments shown in Figures 5 and 6 are used to implement the embodiments in Figures 3 or 4. The specific execution steps and methods of the devices shown in Figures 5 and 6 can refer to the contents of the aforementioned method embodiments.

[0253] Figure 7 is a schematic block diagram of a communication system 700 according to an embodiment of the present application. The communication system 700 includes a first device and a second device, and the first device and the second device are used to implement the embodiment of Figure 3 or Figure 4 above.

[0254] The present application also provides a chip, including a processor, for calling and executing instructions stored in a memory from the memory, so that a communication device equipped with the chip executes the methods in the above examples.

[0255] The present application also provides another chip, comprising: an input interface, an output interface, and a processor, wherein the input interface, the output interface, and the processor are connected via an internal connection path, and the processor is configured to execute code in a memory. When the code is executed, the processor is configured to execute the methods in the above examples. Optionally, the chip also includes a memory, which is configured to store computer programs or code.

[0256] The present application also provides a processor, which is coupled to a memory and is used to execute the methods and functions involving the first device or the second device in any of the above embodiments.

[0257] In another embodiment of the present application, a computer program product including a computer program or instructions is provided. When the computer program product is run on a computer, the method of the aforementioned embodiment is implemented.

[0258] The present application also provides a computer program. When the computer program is executed in a computer, the method of the aforementioned embodiment is implemented.

[0259] In another embodiment of the present application, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a computer, the method described in the above embodiment is implemented.

[0260] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0261] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0262] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0263] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0264] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0265] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a second device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

[0266] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.< / ioc> < / ioc> < / ioc> < / ioc> < / ioc> < / ioc> < / ioc> < / ioc> < / ioc> < / ioc> < / ioc> < / ioc> < / ioc> < / ioc> < / ioc> < / ioc> < / ioc> < / ioc>

Claims

1. A communication method, characterized in that, Including: Receiving first information, where the first information is used to indicate simulating the capabilities of artificial intelligence (AI) / machine learning (ML), and / or simulating the inference types of the AI / ML; Performing a simulation on the AI / ML according to the first information to obtain a simulation report; Sending the simulation report.

2. The method according to claim 1, characterized in that The first information further includes at least one of the following: The model identifier of the AI / ML; The model identifier group of the AI / ML; The activation state, used to indicate turning on or off the simulation of the AI / ML; The simulation area, used to indicate the area where the AI / ML is simulated; The simulation time, used to indicate the time when the AI / ML is simulated; The first request, used to request the simulation result of the capabilities of the AI / ML; The second request, used to request the simulation result of the inference type; The period information, used to indicate the sending period of the simulation report; The performance metrics of the capabilities of the AI / ML; The performance metrics of the inference types of the AI / ML model; The simulation strategy, including at least one of the performance monitoring threshold, performance monitoring area, or performance monitoring time of the AI / ML model; or, The termination strategy, used to indicate the conditions for terminating the simulation of the AI / ML.

3. The method according to claim 1 or 2, characterized in that, The capabilities of the AI / ML include at least one of the following: The capabilities of traffic analysis, coverage analysis, mobility analysis, load analysis, fault analysis, service experience analysis, energy efficiency analysis, or energy consumption analysis.

4. The method according to any one of claims 1 to 3, characterized in that, The inference types of the AI / ML include at least one of the following: The inference type of management data analysis, the inference type of self-optimizing network, the inference type of network data analysis function, or the inference type of radio access network intelligence.

5. The method according to any one of claims 1 to 4, characterized in that, The simulation report includes at least one of the following: The model identifier of the AI / ML; The model identifier group of the AI / ML; The simulation state, used to indicate the state of simulating the AI / ML; The simulation inference capability result, used to indicate the simulation result of the capabilities of the AI / ML; or, The simulation inference type result, used to indicate the simulation result of the inference types of the AI / ML.

6. The method according to claim 5, wherein The simulation inference capability result includes at least one of the following: Traffic congestion problems, burst traffic congestion, non-burst traffic congestion, cell / area average throughput, edge throughput, traffic congestion recovery suggestions, Weak coverage, over-coverage, coverage holes, out-of-cell coverage, coverage problem area information, coverage problem cell identifier, reference signal received power distribution, signal-to-interference-plus-noise ratio distribution, reference signal received quality distribution, coverage optimization suggestions, Handover success rate, handover failure rate, number of premature handovers, number of late handovers, number of handovers to wrong cells, mobility optimization suggestions, Cell physical resource block utilization rate, cell user connection number, central processing unit usage rate, load optimization suggestions, Fault problems, fault problem cell identifier, fault severity, fault recovery suggestions, Energy efficiency problems, energy efficiency problem cell or base station identifier, energy consumption problems, energy consumption problem cell or base station identifier, available output power of the base station, number of users served at a specific power, or energy saving suggestions.

7. The method according to claim 5 or 6, characterized in that, The simulation results of the inference types of the AI / ML include at least one of the following: The identifier of a cell / base station, the information of a deactivated cell / carrier / time slot, the cell-independent bias, the trigger time, or the performance metrics of the inference type.

8. A communication method, characterized in that, Including: Sending a first piece of information, where the first piece of information is used to indicate simulating the capabilities of artificial intelligence AI / machine learning ML and / or simulating the inference types of the AI / ML; Receiving a simulation report.

9. The method according to claim 8, characterized in that The first piece of information further includes at least one of the following: The model identifier of the AI / ML; The group of model identifiers of the AI / ML; The activation status, used to indicate whether to enable or disable the simulation of the AI / ML; The simulation area, used to indicate the area where the AI / ML is simulated; The simulation time, used to indicate the time when the AI / ML is simulated; A first request, used to request the simulation results of the capabilities of the AI / ML; A second request, used to request the simulation results of the inference type; The period information, used to indicate the sending period of the simulation report; The performance metrics of the capabilities of the AI / ML; The performance metrics of the inference types of the AI / ML model; The simulation strategy, including at least one of the performance monitoring threshold, performance monitoring area, or performance monitoring time of the AI / ML model; or, The termination strategy, used to indicate the conditions for terminating the simulation of the AI / ML.

10. The method according to claim 8 or 9, characterized in that, The capabilities of the AI / ML include at least one of the following: The capabilities of traffic analysis, coverage analysis, mobility analysis, load analysis, fault analysis, service experience analysis, energy efficiency analysis, or energy consumption analysis.

11. The method according to any one of claims 8 to 10, characterized in that The inference types of the AI / ML include at least one of the following: The inference type of management data analysis, the inference type of self-optimizing network, the inference type of network data analysis function, or the inference type of radio access network intelligence.

12. The method according to any one of claims 8 to 11, characterized in that The simulation report includes at least one of the following: The model identifier of the AI / ML; The group of model identifiers of the AI / ML; The simulation status, used to indicate the status of simulating the AI / ML; The simulation inference capability result, used to indicate the simulation results of the capabilities of the AI / ML; or, The simulation inference type result, used to indicate the simulation results of the inference types of the AI / ML.

13. The method according to claim 12, wherein The simulation inference capability result includes at least one of the following: Traffic congestion problems, burst traffic congestion, non-burst traffic congestion, cell / area average throughput, edge throughput, traffic congestion recovery suggestions, Weak coverage, over-coverage, coverage holes, out-of-cell coverage, coverage problem area information, coverage problem cell identifier, reference signal received power distribution, signal-to-interference-plus-noise ratio distribution, reference signal received quality distribution, coverage optimization suggestions, Handover success rate, handover failure rate, number of premature handovers, number of late handovers, number of handovers to wrong cells, mobility optimization suggestions, Cell physical resource block utilization, cell user connection number, central processor usage rate, load optimization suggestions, Fault problems, fault problem cell identifier, fault severity, fault recovery suggestions, Energy efficiency issues, cell or base station identifiers for energy efficiency issues, energy consumption issues, cell or base station identifiers for energy consumption issues, available output power of the base station, number of users served at a specific power, or energy-saving suggestions.

14. The method according to claim 12 or 13, characterized in that, The simulation results of the inference type of the AI / ML include at least one of the following: Identifiers of cells / base stations, information on shutting down cells / carriers / time slots, cell-independent biases, trigger times, or performance metrics of the inference type.

15. A communication device, characterized in that, Comprising at least one module, where the at least one module is configured to execute the method according to any one of claims 1 to 7, or the at least one module is configured to execute the method according to any one of claims 8 to 14.

16. A communication device, characterized in that, Comprising a processing circuit and an input / output interface, where the input / output interface is used for inputting and / or outputting signals, and the processing circuit is configured to execute the method according to any one of claims 1 to 7, or the processing circuit is configured to execute the method according to any one of claims 8 to 14.

17. A communication system, characterized in that, Comprising a first device and a second device, where the first device is configured to execute the method according to any one of claims 1 to 7, and the second device is configured to execute the method according to any one of claims 8 to 14.

18. A communication method, characterized in that, The method is applied to the first device and the second device, and the method includes: The first device executes the method according to any one of claims 1 to 7; The second device executes the method according to any one of claims 8 to 14.

19. A computer-readable storage medium, characterized in that, A computer program or instruction is stored on the computer-readable storage medium, and when the computer program or the instruction runs on a computer, the method according to any one of claims 1 to 7 is executed, or the method according to any one of claims 8 to 14 is executed.

20. A computer program product, characterized in that, Comprising a computer program or instruction, and when the computer program or the instruction is run, the method according to any one of claims 1 to 7 is implemented, or the method according to any one of claims 8 to 14 is implemented.

Images