Message passing method and device, network equipment, storage medium and computer program product

Abstract

Disclosed are a message passing method and apparatus, a network device, a storage medium and a computer program product, the method comprising: a first node receiving first information sent by a second node, the first information representing QoS-related information of an AI service; and / or, according to the first information and / or second information, third information is determined, the second information represents available transmission resources and / or time-frequency resources, and the third information represents QoS parameters and / or QoS strategies.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of communication, and particularly relates to a message transmission method and device, network equipment, storage medium and computer program product. BACKGROUND

[0002] In an endogenous artificial intelligence (AI) architecture, an AI computing control node is deployed at an access network (AN) side, and a network exposure function (NEF) of a core network provides quality of service (QoS) information to guarantee the service quality in an AI service implementation process, which increases signaling interaction and link latency between the core network and the AN. SUMMARY

[0003] To solve the problems in the related art, the embodiments of the present application provide a message transmission method and device, network equipment, storage medium and computer program product.

[0004] The technical scheme of the embodiments of the present application is implemented as follows:

[0005] The embodiments of the present application provide a message transmission method applied to a first node, and the method comprises the following steps.

[0006] receiving first information sent by a second node, wherein the first information represents QoS related information of an AI service; and / or,

[0007] determining third information according to the first information and / or second information, wherein the second information represents available transmission resources and / or time-frequency resources, and the third information represents a QoS parameter and / or a QoS policy.

[0008] In the above scheme, the determining of the third information according to the first information and / or the second information comprises the following steps.

[0009] In a case where a set condition is met, the third information is determined according to the first information and / or the second information.

[0010] In the above scheme, the set condition comprises one or more of the following conditions.

[0011] The type of the AI service is a set type;

[0012] The AI service belongs to a latency-sensitive AI service;

[0013] The latency requirement of the AI service is met;

[0014] The accuracy requirement of the AI service is met;

[0015] scheduling transmission resources and / or computing resources for the AI service to meet a latency requirement of the AI service;

[0016] scheduling transmission resources and / or computing resources for the AI service to not meet an accuracy requirement of the AI service.

[0017] In the above solution, the method further includes:

[0018] sending the third information to a network device; or

[0019] providing the AI service to the second node according to the third information.

[0020] In the above solution, the second node includes a terminal, and the method further includes:

[0021] sending a first request to the second node; the first request is used to acquire the first information.

[0022] In the above solution, the first request carries one or more of the following:

[0023] configuration information for reporting the first information;

[0024] reporting time;

[0025] fourth information, the fourth information indicating periodic reporting of the first information or event-triggered reporting of the first information;

[0026] reporting period;

[0027] triggering event.

[0028] In the above solution, the second node includes a terminal and / or an application function (AF).

[0029] In the above solution, the first information includes one or more of the following:

[0030] a first index indicating a requirement of the AI service;

[0031] a latency of the AI service;

[0032] an accuracy of the AI service;

[0033] a first numerical value representing a level of the AI service;

[0034] a first parameter used to assist in generating a QoS parameter and / or a QoS policy;

[0035] a first identifier indicating an importance of the first index.

[0036] In the foregoing solution, the third information indicates one or more of the following:

[0037] The transmission resource does not meet the requirement of the AI service;

[0038] The delay requirement of the AI service is not met;

[0039] The uplink data is compressed;

[0040] The delay requirement of the AI service is met preferentially;

[0041] The accuracy requirement of the AI service is met preferentially;

[0042] The uplink transmission content is reduced;

[0043] The link delay is reduced;

[0044] The link delay is reduced without affecting the accuracy;

[0045] The delay requirement of the AI service is not considered;

[0046] The integrity of all data transmission is ensured;

[0047] Other QoS strategies are requested.

[0048] Embodiments of the present application also provide a message transmission method applied to a second node, and the method comprises the following steps.

[0049] Sending first information to a first node, wherein the first information represents QoS-related information of an AI service.

[0050] In the foregoing solution, the method further comprises the following steps.

[0051] Receiving a first request sent by the first node, wherein the first request is used to acquire the first information.

[0052] In the foregoing solution, the first request carries one or more of the following:

[0053] Configuration information of the first information;

[0054] Reporting time;

[0055] Fourth information, wherein the fourth information indicates periodic reporting of the first information or event-triggered reporting of the first information;

[0056] Reporting period;

[0057] Triggering event.

[0058] In the foregoing solution, the second node comprises a terminal and / or an AF.

[0059] Embodiments of the present application also provide a message transmission device, which comprises the following parts.

[0060] a first receiving unit, configured to receive first information sent by a second node, the first information representing QoS-related information of an AI service; and / or,

[0061] a determining unit, configured to determine third information according to the first information and / or second information, the second information representing available transmission resources and / or time-frequency resources, and the third information representing a QoS parameter and / or a QoS policy.

[0062] Embodiments of the present application further provide a messaging device, comprising:

[0063] a first sending unit, configured to send first information to a first node, the first information representing QoS-related information of an AI service.

[0064] Embodiments of the present application further provide a first node, comprising a first processor and a first communication interface; wherein,

[0065] the first communication interface is configured to receive first information sent by a second node, the first information representing QoS-related information of an AI service; and / or,

[0066] the first processor is configured to determine third information according to the first information and / or second information, the second information representing available transmission resources and / or time-frequency resources, and the third information representing a QoS parameter and / or a QoS policy.

[0067] Embodiments of the present application further provide a second node, comprising a second processor and a second communication interface; wherein,

[0068] the second communication interface is configured to send first information to a first node, the first information representing QoS-related information of an AI service.

[0069] Embodiments of the present application further provide a network device, comprising a processor and a memory for storing a computer program capable of running on the processor,

[0070] wherein the processor is configured to run the computer program to perform the steps of any method on the first node side or the steps of any method on the second node side.

[0071] Embodiments of the present application further provide a storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps of any method on the first node side or the steps of any method on the second node side.

[0072] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of any of the above methods.

[0073] In the message transmission method, apparatus, network device, storage medium, and computer program product provided in the embodiments of this application, a second node sends first information to a first node; the first node receives the first information sent by the second node; and / or, based on the first information and / or the second information, determines third information; the first information represents QoS-related information of the AI ​​service; the second information represents available transmission resources and / or time-frequency resources; and the third information represents QoS parameters and / or QoS policies. In this scheme, the second node can directly provide the first information to the first node to provide service guarantees for the AI ​​service. That is, the first node directly receives the first information, shortening the transmission path of the first information, reducing signaling interaction and link latency during the transmission process, and compared to existing methods for transmitting QoS parameters, reducing signaling complexity and enabling dynamic and rapid guarantee of the service quality of the AI ​​service. Attached Figure Description

[0074] Figure 1 This is a schematic diagram illustrating the QoS process for ensuring services in related technologies.

[0075] Figure 2 This is a schematic flowchart of a message passing method according to an embodiment of this application;

[0076] Figure 3 This is a schematic diagram of a first node deployment architecture according to an embodiment of this application;

[0077] Figure 4 This is a schematic diagram of a first node deployment architecture according to an embodiment of this application;

[0078] Figure 5 This is a schematic flowchart of a third information determination method according to an embodiment of this application;

[0079] Figure 6 This is a schematic flowchart of a message passing method according to an embodiment of this application;

[0080] Figure 7 This is a schematic flowchart of a message passing method according to an application embodiment of this application;

[0081] Figure 8 This is a schematic diagram of a message transmission device according to an embodiment of this application;

[0082] Figure 9 This is a schematic diagram of a message transmission device according to an embodiment of this application;

[0083] Figure 10This is a schematic diagram of the first node structure in an embodiment of this application;

[0084] Figure 11 This is a schematic diagram of the second node structure in an embodiment of this application. Detailed Implementation

[0085] In 5G networks, to ensure service QoS, the Application Filter (AF) sends a QoS request to the Policy Control Function (PCF), carrying QoS reference information. If the AF is a third-party application function, it sends a QoS request to the PCF through network open capabilities. The PCF generates Policy Control and Charging (PCC) rules based on the QoS reference information and sends them to the Session Management Function (SMF). The SMF generates a QoS policy and sends it to the User Plane Function (UPF), the User Interface Controller (AN), and the User Equipment (UE). The network elements receiving the QoS policy jointly execute the QoS policy to ensure service quality.

[0086] Specifically, such as Figure 1As shown, the AF sends a first QoS request to the NEF, for example, the first QoS request could be Nnef_AFSessionWithQoS_Create Request; after authorization by the NEF, the NEF sends a second QoS request carrying QoS reference information to the PCF, for example, the NEF sends a second QoS request to the PCF, for example, the second QoS request could be Npcf_PolioyAuthorization_Create request, or the NEF sends a third QoS request to the Time Sensitive Communication and Time Synchronization Function (TSCTSF), for example, the third QoS request could be Ntsctsf_QoSandTSCAssistance_Create request. The TSCTSF sends a fourth QoS request to the PCF based on the third QoS request, for example, the fourth QoS request could be Npcf_PolicyAuthorization_Update request; the PCF sends a first response to the NEF based on the second QoS request, for example, the first response could be Npcf_PolicyAuthorization_Create. The PCF may send a second response to the TSCTSF based on the fourth QoS request, for example, the second response could be Npcf_PolicyAuthorization_Updateresponse; the TSCTSF may send a third response to the NEF regarding the third QoS request based on the second response, for example, the third response could be Ntsctsf_QoSandTSCAssistance_Create response; the NEF may send a fourth response to the AF regarding the first QoS request based on the first response or the third response, for example, the fourth response could be Nnef_AFsessionWithQos_Createresponse; the NEF may also send a first subscription message to the PCF, for example, the first subscription message could be Npcf_PolicyAuthorization_Subscribe, and the PCF may return a first notification message to the NEF based on the first subscription message, for example, the first notification message could be Npcf_PolicyAuthorization_Notify;The TSCTSF can also send a second subscription message to the PCF, for example, the second subscription message could be `Npcf_PolicyAuthorization_Subscribe`. Based on the second subscription message, the PCF sends a second notification message to the TSCTSF, for example, `Npcf_PolicyAuthorization_Notify`. The TSCTSF sends a third notification message to the NEF, for example, `Ntsctsf_QoSandTSCAssistance_Notify`. The NEF sends a fourth notification message to the AF, for example, `Nnef_AFsessionWithQos_Notify`.

[0087] The AF can also provide individual QoS parameters, which include one or more of the following: requested 5th Generation System (5GS) latency, requested priority, requested Guaranteed Flow Bit Rate (GFBR), requested Maximum Flow Bit Rate, flow direction, burst size, burst arrival time at the UE (uplink), burst arrival time at the UPF (downlink), periodicity, time domain, and lifetime. When the AF provides alternative service requests, it can provide a set of QoS-related parameters for each QoS reference.

[0088] To improve user experience, reduce latency in AI services, and enhance the flexibility of AI service interaction without significant modifications to the terminal, AI applications can be introduced on the AN side, i.e., endogenous AI. In the endogenous AI architecture, AI computing functions are deployed in the base station, and the quality of service during AI service implementation is guaranteed by the base station and terminal sides. Since the core network does not guarantee the service link during service execution, the AI ​​computing control node (orchestration and management of AI applications) can be deployed on the AN side. In this case, QoS information is provided through the NEF of the core network. When adding or modifying QoS parameters, the service link will increase signaling interaction and link latency between the core network and the access network through AF, NEF, PCF, SMF, Access and Mobility Management Function (AMF), and Radio Access Network (RAN).

[0089] Based on this, in various embodiments of this application, the second node sends first information to the first node; the first node receives the first information sent by the second node; and / or, determines third information based on the first information and / or the second information; the first information represents QoS-related information of the AI ​​service; the second information represents available transmission resources and / or time-frequency resources; and the third information represents QoS parameters and / or QoS policies. In the above scheme, the second node can directly provide the first information to the first node to provide service guarantees for the AI ​​service. That is, the first node directly receives the first information, shortening the transmission path of the first information, reducing signaling interaction and link latency during the transmission process, and compared with existing methods for transmitting QoS parameters, reducing signaling complexity and enabling dynamic and rapid guarantee of the service quality of the AI ​​service.

[0090] The present application will now be described in further detail with reference to the accompanying drawings and embodiments.

[0091] This application provides a message passing method applied to a first node. The first node can be deployed independently on the radio access network side or the access network side, or it can be co-located with network equipment, including base stations and access network equipment. The first node is used to receive and / or process first information. The first node can also be referred to as RAN-E (Radio Access Network–Extend). Figure 2 As shown, the method includes:

[0092] Step 201: Receive the first information sent by the second node; and / or,

[0093] Step 202: Determine the third information based on the first information and / or the second information.

[0094] The first information represents QoS-related information of the AI ​​service; the second information represents available transmission resources and / or time-frequency resources; and the third information represents QoS parameters and / or QoS policies.

[0095] Here, the first node can be a newly added node, independently deployed on the radio side or the access network side. When the first node is deployed independently, it can receive first information sent by the second node and determine third information based on the first and / or second information; alternatively, the first node can receive the first information sent by the second node and forward it to the network device, so that the network device can determine the third information based on the first and / or second information. The network device can be a base station, RAN, RAN equipment, or access network equipment. For example, ... Figure 3As shown, a first node (wireless enhancement network element) is added outside the base station or RAN to implement wireless enhancement functions, receive first information, and interact with the RAN. In the architecture where the first node is deployed independently, the first node interacts with network devices, which guarantee the quality of service for AI services. That is, the ordinary service data links are still UE, RAN, core network (CN), and data network (DN), which makes the structure of the independent deployment of the first node highly compatible with backward compatibility and does not affect the overall network architecture.

[0096] The first node can also be co-located with network devices. It can be deployed within the network device's protocol stack. The network device can receive first information sent by the second node and determine third information based on the first and / or second information. Specifically, the first node can be deployed at the protocol layer of the network device, such as the Radio Resource Control (RRC) layer or the Non-Access Stratum (NAS) layer. The quality of service for the AI ​​service is guaranteed by the execution function modules within the network device. For example, ... Figure 4 As shown, the first node is deployed in the RRC layer of the base station. The RRC layer is responsible for receiving and processing the first information. Since the RRC layer is deployed in the Centralized Unit (CU), the base station manufacturer only needs to optimize the CU. The base station also deploys the Packet Data Convergence Protocol (PDCP) layer, the Radio Link Control (RLC) layer, the Media Access Control (MAC) layer, and the Physical (PHY) layer.

[0097] When the first node is deployed within the protocol stack of the network device, the function of receiving and processing the first information can be realized by optimizing the protocol stack. In a network architecture where the CU and the Distributed Unit (DU) are deployed separately, the vendor only needs to optimize its own equipment, which reduces the number of new node connections compared to the case where the first node is deployed independently.

[0098] It should be noted that AI services can represent the network providing AI technology, AI services, or the three elements of AI to the service requester (user) on demand. AI services can also be called AI applications. The second node can be a terminal or AF, and the terminal can also be called UE or terminal device. The first node has network openness and can communicate with the UE and CN; the first node may need CN authorization to realize the function of receiving first information.

[0099] In this embodiment, the interaction between the first node and the second node is directly reduced, which reduces the signaling interaction and link latency between the core network and the access network during the first information transmission process. Furthermore, it reduces the signaling complexity for the transmission of traditional QoS parameters in the AI ​​service of the third-party AF.

[0100] In order to provide service guarantees based on the characteristics of AI services and improve resource utilization, in one embodiment, the first information includes one or more of the following:

[0101] The first indicator, which indicates the demand for AI services;

[0102] Latency of AI services;

[0103] The accuracy of AI services;

[0104] The first value represents the level of the AI ​​service;

[0105] The first parameter is used to assist in generating QoS parameters and / or QoS policies;

[0106] The first identifier indicates the importance of the first indicator.

[0107] Here, the first information may include at least one of a first indicator, a first value, and a first parameter; wherein, the first indicator may include the latency of the AI ​​service and / or the accuracy of the AI ​​service; additionally, when the first information includes two or more first indicators, for example, the first information includes the latency of the AI ​​service and the accuracy of the AI ​​service, the first information may also include a first identifier to identify the indicator that must be prioritized among the first indicators, for example, identifying the indicator that must be prioritized between the latency of the AI ​​service and the accuracy of the AI ​​service. The first information may also include the identifier of the AI ​​service and / or the type of the AI ​​service.

[0108] The first indicator can also be understood as an indicator related to AI service quality assurance, or as an indicator representing user AI service needs from the user's perspective. AI service latency can be understood as end-to-end AI service latency, which can include computation latency and transmission latency (also called connection latency). AI service accuracy can include absolute accuracy and relative accuracy. The first value can be understood as an identifier of the AI ​​service level, or as an AI Service Key Indicator (AISKI). There is a mapping relationship between the first value and the first indicator. The first parameter can be understood as quantifiable terminal and channel parameters, such as terminal computing power, available terminal computing power, channel quality, and terminal battery power.

[0109] It should be noted that the first information may contain two or more mutually restrictive first indicators. For example, the first information may include the latency and accuracy of the AI ​​service. Since network resources are limited, quantizing or compressing transmitted content to ensure the latency of the AI ​​service will affect its accuracy; that is, the latency and accuracy of the AI ​​service are mutually restrictive. Therefore, the first information also needs to include a first identifier to clarify the importance of the mutually restrictive first indicators, so that when available resources cannot simultaneously meet all first indicators, the first indicator that needs to be prioritized can be selected. For example, the first identifier can be used to classify the first indicators into important and unimportant categories, or the first identifier can indicate the importance level of each first indicator. For instance, if the first indicators include the latency and accuracy of the AI ​​service, and the first identifier for the latency is important while the first identifier for the accuracy is unimportant, then the fact that the first identifier for the latency is important can be understood as the AI ​​service type being latency-critical. For example, if the first indicator includes the latency or accuracy of the AI ​​service, it can be understood that the type of AI service corresponds to the first indicator, and is either latency-critical or accuracy-critical.

[0110] To flexibly ensure the service quality of AI services, in one embodiment, determining the third information based on the first and / or second information includes:

[0111] If the set conditions are met, the third information is determined based on the first information and / or the second information.

[0112] Here, it can be determined whether the set conditions are met based on the first information and / or the second information; if the set conditions are met, the third information is determined based on the first information and / or the second information; if the set conditions are not met, the first information is ignored, and it is not necessary to determine the third information based on the first information. The set conditions can be set according to actual needs.

[0113] In order to improve resource utilization while ensuring the quality of AI services, the set conditions include one or more of the following:

[0114] The type of AI service is set to a specific type;

[0115] The AI ​​service is a latency-sensitive AI service;

[0116] To meet the latency requirements of AI services;

[0117] To meet the accuracy requirements of AI services;

[0118] Orchestrate transmission and / or computing resources for AI services to meet the latency requirements of AI services;

[0119] The transmission and / or computing resources allocated to the AI ​​service do not meet the accuracy requirements of the AI ​​service.

[0120] Here, the type can be configured to be latency-critical and / or accuracy-critical. The type of AI service can be determined based on the AI ​​service's identifier.

[0121] When the configuration type is latency-critical, the available transmission resources and / or time-frequency resources must prioritize meeting the latency requirements of the AI ​​service, i.e., the first indicator indicating the latency requirements of the AI ​​service must be prioritized. In this case, the configuration conditions may include any of the following: the AI ​​service type is the configuration type, the AI ​​service belongs to the latency-sensitive category, or the latency requirements of the AI ​​service are met. The first node or network device can determine third information based on the first and / or second information to meet the latency requirements of the AI ​​service. The first information can be used to allocate or orchestrate transmission resources and / or computing resources for the AI ​​service. The configuration conditions may also include orchestrating transmission resources and / or computing resources for the AI ​​service to meet its latency requirements. Transmission resources include time-domain resources and / or frequency-domain resources.

[0122] When the setting condition is accuracy-critical, the available transmission resources and / or time-frequency resources must prioritize meeting the accuracy requirements of the AI ​​service. That is, the first indicator indicating the accuracy requirements of the AI ​​service must be prioritized. In this case, the setting condition includes the type of AI service being the set type and / or meeting the accuracy requirements of the AI ​​service. The first node or network device can determine third information based on the first and / or second information to meet the accuracy requirements of the AI ​​service. The setting condition may also include transmission resources and / or computing resources allocated to the AI ​​service that do not meet the accuracy requirements of the AI ​​service.

[0123] To fully utilize available resources and dynamically and quickly ensure the service quality of AI services, in one embodiment, the third information indicates one or more of the following:

[0124] The transmission resources do not meet the needs of AI services;

[0125] The latency requirements for AI services are not met.

[0126] Compress upstream data;

[0127] Prioritize meeting the latency requirements of AI services;

[0128] Prioritize meeting the accuracy requirements of AI services;

[0129] Reduce the amount of data transmitted upstream;

[0130] Reduce link latency;

[0131] Reduce link latency without affecting accuracy;

[0132] Latency requirements for AI services are not taken into account;

[0133] Ensure the integrity of all data transmissions;

[0134] Request alternative QoS policies.

[0135] Here, when the type is set to latency-critical, the third information may include one or more of the following: transmission resources do not meet the needs of AI services, uplink data is compressed, latency requirements of AI services are prioritized, uplink transmission content is reduced, link latency is reduced, link latency is reduced without affecting accuracy, latency requirements of AI services are prioritized, and transmission resources do not meet the needs of AI services.

[0136] When the setting condition is accuracy-critical, the third information may include one or more of the following: transmission resources do not meet the needs of AI services, do not meet the latency requirements of AI services, prioritize meeting the accuracy requirements of AI services, disregard the latency requirements of AI services, ensure the integrity of all data transmissions, request other QoS policies, and prioritize meeting the accuracy requirements of AI services.

[0137] It should be noted that if the transmission resources orchestrated for the AI ​​service do not meet the accuracy requirements of the AI ​​service, the latency requirements of the AI ​​service can be disregarded to ensure that the transmission resources orchestrated for the AI ​​service can guarantee the integrity of all data transmissions. If the computing resources orchestrated for the AI ​​service do not meet the accuracy requirements of the AI ​​service, other QoS policies can be requested. If both the transmission and computing resources orchestrated for the AI ​​service do not meet the accuracy requirements of the AI ​​service, any of the above-mentioned solutions can be selected, or a combination of the two solutions can be used to determine the third information.

[0138] For example, the process for determining third-party information is as follows: Figure 5 As shown, the details are as follows:

[0139] Step 501: Obtain the first information and / or the second information.

[0140] The specific implementation of step 501 is described above and will not be repeated here.

[0141] Step 502: Determine whether the type of AI service is the set type.

[0142] Here, as Figure 5As shown, the type can be set to latency-critical; if the AI ​​service type is not latency-critical, or if the AI ​​service type is accuracy-critical, proceed to step 503; if the AI ​​service type is latency-critical, proceed to step 507.

[0143] Step 503: Determine whether the transmission resources and / or computing resources orchestrated for the AI ​​service meet the accuracy requirements of the AI ​​service.

[0144] Here, based on the first information and / or the second information, transmission resources and / or computing resources can be allocated or orchestrated for the AI ​​service, and it can be determined whether the transmission resources and / or computing resources orchestrated for the AI ​​service meet the accuracy requirements of the AI ​​service.

[0145] If the transmission and / or computing resources allocated to the AI ​​service meet the accuracy requirements of the AI ​​service, step 509 can be executed directly to determine the third information, or step 507 can be executed; if the transmission and / or computing resources allocated to the AI ​​service cannot meet the accuracy requirements of the AI ​​service, step 504 is executed to further locate the reason why the accuracy requirements of the AI ​​service cannot be met.

[0146] Step 504: Determine whether the transmission resources orchestrated for the AI ​​service meet the accuracy requirements of the AI ​​service.

[0147] Here, if the transmission resources orchestrated for the AI ​​service do not meet the accuracy requirements of the AI ​​service, step 505 is executed to schedule available resources to meet the accuracy requirements of the AI ​​service; if the transmission resources orchestrated for the AI ​​service meet the accuracy requirements of the AI ​​service, it indicates that other reasons may be causing the inability to meet the accuracy requirements of the AI ​​service, such as the computing resources orchestrated for the AI ​​service not meeting the accuracy requirements of the AI ​​service, and step 506 is executed.

[0148] Step 505: Ignore the latency requirements of AI services and ensure the integrity of all data transmissions.

[0149] Here, since the AI ​​service type is accuracy-critical, meaning that the accuracy requirements of the AI ​​service need to be met first, the latency requirements of the AI ​​service can be ignored to ensure the integrity of all data transmission, thereby enabling the transmission resources orchestrated for the AI ​​service to meet the accuracy requirements of the AI ​​service.

[0150] Step 506: Request other QoS policies.

[0151] Here, other QoS strategies could be to schedule and orchestrate computing resources allocated to AI services so that the computing resources allocated to AI services meet the accuracy requirements of AI services; for example, suspending other services or businesses with lower AI service levels; or querying the remaining usage time of computing resources for other services or businesses that consume computing resources, and then providing AI services after the other services or businesses that consume computing resources have finished, so that the computing resources allocated to AI services meet the accuracy requirements of AI services.

[0152] Step 507: Determine whether the transmission resources and / or computing resources orchestrated for the AI ​​service meet the latency requirements of the AI ​​service.

[0153] Here, if the transmission and / or computing resources orchestrated for the AI ​​service meet the latency requirements of the AI ​​service, step 509 is executed; if the transmission and / or computing resources orchestrated for the AI ​​service cannot meet the latency requirements of the AI ​​service, step 508 is executed.

[0154] Step 508: Reduce uplink transmission content and / or, without affecting accuracy, reduce link latency.

[0155] Here, the third piece of information indicates reducing uplink transmission content and / or, without affecting accuracy, reducing link latency. Since the AI ​​service type is latency-critical, meaning that latency requirements for the AI ​​service must be prioritized, link latency can be reduced as much as possible without affecting accuracy. Reducing link latency can be achieved by reducing uplink transmission content; for example, this can be done through quantization or compression to reduce the amount of data transmitted, thereby ensuring that the transmission and / or computing resources orchestrated for the AI ​​service can meet its latency requirements.

[0156] Step 509: Output the third information.

[0157] Here, if the transmission and / or computing resources allocated to the AI ​​service meet the accuracy or latency requirements of the AI ​​service, the transmission and / or computing resources can be allocated directly according to the accuracy or latency requirements of the AI ​​service to provide the AI ​​service; or, according to step 505, step 506, or step 508, the third information can be determined and output.

[0158] To flexibly ensure the service quality of AI services for different deployment methods of the first node, in one embodiment, the method further includes:

[0159] Send the third information to the network device; or,

[0160] The AI ​​service is provided to the second node based on the third information.

[0161] Here, with the first node deployed independently, since the quality of service for AI services is ensured by network devices, the first node needs to send the third information to the network devices after determining the third information. This allows the network devices to ensure the quality of service for AI services based on the third information, thereby reducing signaling interaction and link latency between the core network and the access network without changing the relevant network architecture.

[0162] When the first node is co-located with network equipment, the first node can guarantee the quality of AI services. Therefore, after determining the third information, the first node can directly provide AI services to the second node based on the third information. Compared with the case of deploying the first node independently, this reduces the number of new node connections and reduces the signaling interaction and link latency between the core network and the access network.

[0163] The first node can receive first information proactively reported by the second node, or request to report first information from the second node, or subscribe to first information from the second node. Based on the first node requesting the second node to report first information, or the first node subscribing to first information from the second node, in one embodiment, the second node includes a terminal, and the method further includes:

[0164] Send a first request to the second node; the first request is used to obtain the first information.

[0165] Here, when the first node is deployed independently, the first node can send the first request to the second node through the network device; when the first node is co-located with the network device, such as when the first node is co-located with the base station, the first node can directly send the first request to the second node.

[0166] It should be noted that when the second node includes a terminal, there is frequent data interaction between the first node and the second node. The first node needs to perceive relevant information about the terminal in real time (such as first information, the terminal's available computing power, channel quality, and terminal battery level) in order to coordinate and update the QoS guarantee of the AI ​​service requested by the terminal.

[0167] To save signaling overhead, in one embodiment, the first request carries one or more of the following:

[0168] The configuration information for reporting the first piece of information;

[0169] Reporting time;

[0170] The fourth information indicates that the first information is reported periodically or triggered by an event.

[0171] Reporting cycle;

[0172] Trigger the event.

[0173] Here, the first request may carry configuration information for reporting the first information, instructing the second node to report the first information according to the configuration information. The first request may also carry one or more of the following: reporting time, fourth information, reporting period, and triggering event, to indicate when the second node reports the first information; wherein, when the first request carries a reporting time, the second node reports the first information according to the reporting time; when the first request carries a reporting period or the fourth information indicates a reporting period, the second node reports the first information periodically; when the first request carries a triggering event, or the fourth information indicates that the first information is triggered by an event, the second node reports the first information when it detects that a triggering event has occurred.

[0174] The first information can be transmitted from the terminal to the base station via uplink, or it can be sent by the AF. Based on this, in one embodiment, the second node includes the terminal and / or the AF.

[0175] Here, when the first information is transmitted from the terminal to the first node (i.e., the second node includes the terminal), the terminal can proactively report the first information to the first node, or it can report the first information to the first node upon receiving a first request.

[0176] In the case where the second node includes AF, if the AI ​​service does not have high latency requirements, such as the training service, which aims to obtain a more accurate and generalizable model, and the model used by the training service is authorized to the network device by a third party and can be trained and updated using user data, the second node can proactively send the first information to the first node without the first node needing to request the first information, thereby reducing signaling overhead.

[0177] Correspondingly, this application also provides a message passing method applied to a second node, which may be a node requesting AI services, such as... Figure 6 As shown, the method includes:

[0178] Step 601: Send the first information to the first node. The first information represents QoS-related information of the AI ​​service.

[0179] Here, first information can be generated and sent to the first node based on the characteristics of the AI ​​service, or based on the characteristics of the AI ​​service and relevant information of the second node; or first information can be sent to the first node based on the second node's demand for the AI ​​service, and / or changes in the QoS-related information of the AI ​​service.

[0180] In order to flexibly provide first information based on different AI service requesters, in one embodiment, the second node includes a terminal and / or an AF.

[0181] Here, when the second node includes an AF (Automatic AF), first information is generated based on the characteristics of the AI ​​service and sent to the first node; subsequently, updated first information can be fed back to the first node based on changes in QoS-related information of the AI ​​service. When the second node includes a terminal, first information is generated based on the characteristics of the AI ​​service and the resource status of the second node, and sent to the first node; subsequently, updated first information can be fed back to the first node in a timely manner based on the second node's demand for the AI ​​service and the received first request.

[0182] When the second node includes a terminal, it needs to promptly report the terminal's relevant resources and the latest information to the first node so that the first node can quickly and dynamically ensure the service quality of the AI ​​service. Based on this, in one embodiment, the method further includes:

[0183] Receive a first request sent by the first node, the first request being used to obtain the first information.

[0184] Here, when the second node includes a terminal, it receives a first request sent by the first node, and promptly feeds back first information to the first node based on the first request and relevant information of the second node (such as the available computing power of the terminal, channel quality, and terminal battery level).

[0185] To save signaling overhead, in one embodiment, the first request carries one or more of the following:

[0186] The configuration information for reporting the first piece of information;

[0187] Reporting time;

[0188] The fourth information indicates that the first information is reported periodically or triggered by an event.

[0189] Reporting cycle;

[0190] Trigger the event.

[0191] Here, if the first request carries configuration information for reporting the first information, the second node reports the first information according to that configuration information. If the first request carries a reporting time, the second node reports the first information according to the reporting time. If the first request carries a reporting period or a fourth message indicating a reporting period, the second node reports the first information periodically. If the first request carries a triggering event, or a fourth message indicating that the first information is triggered by an event, the second node reports the first information upon detecting that a triggering event has occurred.

[0192] The following section provides a more detailed description of this application with application examples. This application can be applied to business scenarios that provide AI services on the wireless side, such as mobile robots, autonomous or assisted driving. The message transmission process is as follows: Figure 7 As shown:

[0193] Step 1: The first node sends the first request to the second node.

[0194] Here, when the second node includes a terminal, the first node sends a first request to the second node to instruct the second node to report first information. When the first node is an AF (Automatic Message Passage), the first node does not need to execute step 1; that is, the message passing process does not include step 1 described above.

[0195] Step 2: The second node generates the first information based on the characteristics of the AI ​​service and / or the relevant information of the second node.

[0196] Here, when the second node includes a terminal, the second node generates first information based on the characteristics of the AI ​​service and relevant information of the second node. The relevant information of the second node may include one or more of the terminal's available computing power, channel quality, and terminal battery level. When the second node includes an AF (Automatic Feedback Provider), the second node generates first information based on the characteristics of the AI ​​service. For example, if the second node is a terminal, the first information includes the AI ​​service latency T1 and a Buffer Status Report (BSR), where the AI ​​service latency T1 is 10ms.

[0197] Step 3: The second node sends the first message to the first node.

[0198] The specific implementation methods for steps 3 to 5 are described above and will not be repeated here.

[0199] Step 4: The first node receives the first message sent by the second node.

[0200] Step 5: The first node determines the third information based on the first information and / or the second information.

[0201] Here, for example, the first information includes the latency T1 and BSR of the AI ​​service. The latency T1 of the AI ​​service is 10ms. Based on the first information, transmission resources and / or computing resources are allocated or orchestrated for the AI ​​service, resulting in a transmission latency T2 and a computing latency T3; where the transmission latency T2 is 8ms and the computing latency T3 is 5ms. Since T2 + T3 > T1, the transmission resources and / or computing resources orchestrated for the AI ​​service do not meet the latency requirements of the AI ​​service.

[0202] Step 6: The first node sends the third information to the network device.

[0203] Here, the first node sends the third piece of information to the network device after determining the third piece of information. For example, the third piece of information may include the failure to meet the latency requirements of the AI ​​service.

[0204] Step 7: The network device provides AI services to the second node based on the third information.

[0205] Here, the network device provides AI services to the second node based on the third information. That is, the transmission path of information related to ensuring the quality of service of the AI ​​service only passes through the first node, the second node, and the network device. This reduces signaling interaction and link latency in transmitting information related to ensuring the quality of service of the AI ​​service, lowers signaling complexity, and improves resource utilization. For example, the third information includes the failure to meet the latency requirements of the AI ​​service. Based on the third information, uplink data is compressed, i.e., a compression identifier is generated. An uplink scheduling grant (UL grant) carrying the compression identifier is sent to the second node to instruct the second node to upload the compressed AI service-related data, thereby ensuring the quality of service of the AI ​​service. Here, UL refers to the uplink; in the case of an inference service, the AI ​​service-related data is AI inference data.

[0206] It should be noted that when the first node is deployed independently and the second node is a terminal, the message transmission process includes steps 1 to 7 above; when the first node is co-located with network equipment and the second node is a terminal, the message transmission process does not include step 6 above. After step 7, based on the needs of the second node and / or changes in the QoS-related information of the AI ​​service, the second node promptly feeds back new first information, that is, executes steps 2 to 7 or 2 to 5, 7.

[0207] It should be noted that when the first node is deployed independently and the second node is an AF (Automatic AF), the message passing process includes steps 2 to 7 above, but does not include step 1 above; when the first node is co-located with the network device and the second node is an AF, the message passing process includes steps 2 to 5 and 7 above. After step 7, based on the changes in QoS-related information of the AI ​​service, updated first information is fed back to the first node, i.e., steps 2 to 7 or 2 to 5 and 7 are executed.

[0208] To implement the method on the first node side of this application embodiment, this application embodiment also provides a message passing device, which is disposed on the first node, such as... Figure 8 As shown, the device includes:

[0209] The first receiving unit 801 is configured to receive first information sent by the second node, wherein the first information represents QoS-related information of the AI ​​service; and / or

[0210] The determining unit 802 is configured to determine third information based on the first information and / or the second information, wherein the second information represents available transmission resources and / or time-frequency resources, and the third information represents QoS parameters and / or QoS policies.

[0211] In one embodiment, the determining unit 802 is specifically used to determine third information based on the first information and / or the second information when the set conditions are met.

[0212] In one embodiment, the setting conditions include one or more of the following:

[0213] The type of AI service is set to a specific type;

[0214] The AI ​​service is a latency-sensitive AI service;

[0215] To meet the latency requirements of AI services;

[0216] To meet the accuracy requirements of AI services;

[0217] Orchestrate transmission and / or computing resources for AI services to meet the latency requirements of AI services;

[0218] The transmission and / or computing resources allocated to the AI ​​service do not meet the accuracy requirements of the AI ​​service.

[0219] In one embodiment, the device further includes:

[0220] The second sending unit is used to send the third information to the network device; or...

[0221] The service unit is used to provide the AI ​​service to the second node based on the third information.

[0222] In one embodiment, the second node includes a terminal, and the device further includes:

[0223] The third sending unit is used to send a first request to the second node; the first request is used to obtain the first information.

[0224] In one embodiment, the first request carries one or more of the following:

[0225] The configuration information for reporting the first piece of information;

[0226] Reporting time;

[0227] The fourth information indicates that the first information is reported periodically or triggered by an event.

[0228] Reporting cycle;

[0229] Trigger the event.

[0230] In one embodiment, the second node includes a terminal and / or an AF.

[0231] In one embodiment, the first information includes one or more of the following:

[0232] The first indicator, which indicates the demand for AI services;

[0233] Latency of AI services;

[0234] The accuracy of AI services;

[0235] The first value represents the level of the AI ​​service;

[0236] The first parameter is used to assist in generating QoS parameters and / or QoS policies;

[0237] The first identifier indicates the importance of the first indicator.

[0238] In one embodiment, the third information indicates one or more of the following:

[0239] The transmission resources do not meet the needs of AI services;

[0240] The latency requirements for AI services are not met.

[0241] Compress upstream data;

[0242] Prioritize meeting the latency requirements of AI services;

[0243] Prioritize meeting the accuracy requirements of AI services;

[0244] Reduce the amount of data transmitted upstream;

[0245] Reduce link latency;

[0246] Reduce link latency without affecting accuracy;

[0247] Latency requirements for AI services are not taken into account;

[0248] Ensure the integrity of all data transmissions;

[0249] Request alternative QoS policies.

[0250] In practical applications, the first receiving unit 801, the second sending unit, and the third sending unit can be implemented by a processor in the message passing device combined with a communication interface, and the determining unit 802 and the service unit can be implemented by a processor in the message passing device.

[0251] To implement the method on the second node side of this application embodiment, this application embodiment also provides a message passing device, disposed on the second node, such as... Figure 9 As shown, the device includes:

[0252] The first sending unit 901 is used to send first information to the first node, wherein the first information represents QoS-related information of the AI ​​service.

[0253] In one embodiment, the device further includes:

[0254] The second receiving unit is used to receive a first request sent by the first node, the first request being used to obtain the first information.

[0255] In one embodiment, the first request carries one or more of the following:

[0256] The configuration information for reporting the first piece of information;

[0257] Reporting time;

[0258] The fourth information indicates that the first information is reported periodically or triggered by an event.

[0259] Reporting cycle;

[0260] Trigger the event.

[0261] In one embodiment, the second node includes a terminal and / or an AF.

[0262] In practical applications, the first sending unit 901 and the second receiving unit can be implemented by a processor in the message passing device combined with a communication interface.

[0263] It should be noted that the message passing device provided in the above embodiments is only illustrated by the division of the above program modules. In practical applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the device can be divided into different program modules to complete all or part of the processing described above. In addition, the message passing device and message passing method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0264] Based on the hardware implementation of the above program modules, and in order to implement the method on the first node side of the embodiments of this application, the embodiments of this application also provide a first node, such as... Figure 10 As shown, the first node 1000 includes:

[0265] The first communication interface 1001 can exchange information with other network nodes.

[0266] The first processor 1002 is connected to the first communication interface 1001 to enable information interaction with other network nodes and to execute the methods provided by one or more technical solutions on the first node side when running a computer program.

[0267] The first memory 1003 is used to store computer programs that can run on the first processor 1002.

[0268] Specifically, the first communication interface 1001 is used to receive first information sent by the second node, wherein the first information represents QoS-related information of the AI ​​service; and / or,

[0269] The first processor 1002 is configured to determine third information based on the first information and / or the second information, wherein the second information represents available transmission resources and / or time-frequency resources, and the third information represents QoS parameters and / or QoS policies.

[0270] In one embodiment, the first processor 1002 is specifically configured to determine third information based on the first information and / or the second information when a set condition is met.

[0271] In one embodiment, the setting conditions include one or more of the following:

[0272] The type of AI service is set to a specific type;

[0273] The AI ​​service is a latency-sensitive AI service;

[0274] To meet the latency requirements of AI services;

[0275] To meet the accuracy requirements of AI services;

[0276] Orchestrate transmission and / or computing resources for AI services to meet the latency requirements of AI services;

[0277] The transmission and / or computing resources allocated to the AI ​​service do not meet the accuracy requirements of the AI ​​service.

[0278] In one embodiment, the first communication interface 1001 is further configured to send the third information to the network device; or,

[0279] The first processor 1002 is also configured to provide the AI ​​service to the second node based on the third information.

[0280] In one embodiment, the second node includes a terminal, and the first communication interface 1001 is further configured to send a first request to the second node; the first request is used to obtain the first information.

[0281] In one embodiment, the first request carries one or more of the following:

[0282] The configuration information for reporting the first piece of information;

[0283] Reporting time;

[0284] The fourth information indicates that the first information is reported periodically or triggered by an event.

[0285] Reporting cycle;

[0286] Trigger the event.

[0287] In one embodiment, the second node includes a terminal and / or an AF.

[0288] In one embodiment, the first information includes one or more of the following:

[0289] The first indicator, which indicates the demand for AI services;

[0290] Latency of AI services;

[0291] The accuracy of AI services;

[0292] The first value represents the level of the AI ​​service;

[0293] The first parameter is used to assist in generating QoS parameters and / or QoS policies;

[0294] The first identifier indicates the importance of the first indicator.

[0295] In one embodiment, the third information indicates one or more of the following:

[0296] The transmission resources do not meet the needs of AI services;

[0297] The latency requirements for AI services are not met.

[0298] Compress upstream data;

[0299] Prioritize meeting the latency requirements of AI services;

[0300] Prioritize meeting the accuracy requirements of AI services;

[0301] Reduce the amount of data transmitted upstream;

[0302] Reduce link latency;

[0303] Reduce link latency without affecting accuracy;

[0304] Latency requirements for AI services are not taken into account;

[0305] Ensure the integrity of all data transmissions;

[0306] Request alternative QoS policies.

[0307] It should be noted that the specific processing procedures of the first processor 1002 and the first communication interface 1001 can be understood by referring to the above method.

[0308] Of course, in practical applications, the various components in the first node 1000 are coupled together through the bus system 1004. It can be understood that the bus system 1004 is used to implement communication between these components. In addition to the data bus, the bus system 1004 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 10 The general labeled all buses as Bus System 1004.

[0309] The first memory 1003 in this embodiment is used to store various types of data to support the operation of the first node 1000. Examples of such data include any computer program used to operate on the first node 1000.

[0310] The methods disclosed in the above embodiments of this application can be applied to the first processor 1002, or implemented by the first processor 1002. The first processor 1002 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware or by instructions in the form of software in the first processor 1002. The first processor 1002 may be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The first processor 1002 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the methods disclosed in the embodiments of this application can be directly manifested as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium, which is located in the first memory 1003. The first processor 1002 reads the information in the first memory 1003 and completes the steps of the aforementioned method in combination with its hardware.

[0311] In an exemplary embodiment, the first node 1000 may be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers (MCUs), microprocessors, or other electronic components to perform the aforementioned method.

[0312] Based on the hardware implementation of the above program modules, and in order to implement the method on the second node side of the embodiments of this application, the embodiments of this application also provide a second node, such as... Figure 11 As shown, the second node 1100 includes:

[0313] The second communication interface 1101 can exchange information with other network nodes.

[0314] The second processor 1102 is connected to the second communication interface 1101 to enable information interaction with other network nodes and to execute the methods provided by one or more technical solutions on the second node side when running computer programs.

[0315] The second memory 1103 is used to store computer programs that can run on the second processor 1102.

[0316] Specifically, the second communication interface 1101 is used to send first information to the first node, the first information representing QoS-related information of the AI ​​service.

[0317] In one embodiment, the second communication interface 1101 is further configured to receive a first request sent by the first node, the first request being used to obtain the first information.

[0318] In one embodiment, the first request carries one or more of the following:

[0319] The configuration information for reporting the first piece of information;

[0320] Reporting time;

[0321] The fourth information indicates that the first information is reported periodically or triggered by an event.

[0322] Reporting cycle;

[0323] Trigger the event.

[0324] In one embodiment, the second node includes a terminal and / or an AF.

[0325] It should be noted that the specific processing procedures of the second processor 1102 and the second communication interface 1101 can be understood by referring to the above method.

[0326] Of course, in practical applications, the various components in the second node 1100 are coupled together through the bus system 1104. It can be understood that the bus system 1104 is used to implement communication between these components. In addition to the data bus, the bus system 1104 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 11 The general designated all buses as Bus System 1104.

[0327] The second memory 1103 in this embodiment is used to store various types of data to support the operation of the second node 1100. Examples of such data include any computer program used to operate on the second node 1100.

[0328] The methods disclosed in the above embodiments of this application can be applied to the second processor 1102, or implemented by the second processor 1102. The second processor 1102 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware or by instructions in the form of software in the second processor 1102. The second processor 1102 may be a general-purpose processor, a DSP, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The second processor 1102 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the methods disclosed in the embodiments of this application can be directly manifested as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium, which is located in the second memory 1103. The second processor 1102 reads the information in the second memory 1103 and completes the steps of the aforementioned method in conjunction with its hardware.

[0329] In an exemplary embodiment, the second node 1100 may be implemented by one or more ASICs, DSPs, PLDs, CPLDs, FPGAs, general-purpose processors, controllers, MCUs, microprocessors, or other electronic components to perform the aforementioned method.

[0330] It is understood that the memories (first memory 1003 and second memory 1103) in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile memory and non-volatile memory. Specifically, non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM).The memories described in the embodiments of this application are intended to include, but are not limited to, these and any other suitable types of memories.

[0331] In an exemplary embodiment, this application also provides a storage medium, namely a computer storage medium, specifically a computer-readable storage medium, such as a first memory 1003 storing a computer program, which can be executed by a first processor 1002 of a first node 1000 to complete the steps described in the aforementioned first node-side method. Another example is a second memory 1103 storing a computer program, which can be executed by a second processor 1102 of a second node 1100 to complete the steps described in the aforementioned second node-side method. The computer-readable storage medium can be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disc, or CD-ROM.

[0332] For example, this application also provides a computer program product, including a computer program that can be executed by a first processor 1002 of a first node 1000 and a second processor 1102 of a second node 1100 to perform the steps described in any of the foregoing methods.

[0333] It should be noted that "first," "second," etc., are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence; "multiple" refers to two or more items.

[0334] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A exists alone, A and B exist simultaneously, and B exists alone. Additionally, the terms "at least one" or "at least one item" in this document refer to any combination of at least two of any one or more of a plurality of elements. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0335] Furthermore, the technical solutions described in the embodiments of this application can be combined arbitrarily without conflict.

[0336] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application.

Claims

1. A message passing method, characterized in that, Applied to the first node, the method includes: Receive first information sent by the second node, wherein the first information represents QoS-related information of the artificial intelligence (AI) service; and / or, Based on the first information and / or the second information, a third information is determined, wherein the second information represents available transmission resources and / or time-frequency resources, and the third information represents QoS parameters and / or QoS policies.

2. The method according to claim 1, characterized in that, The step of determining the third information based on the first information and / or the second information includes: If the set conditions are met, the third information is determined based on the first information and / or the second information.

3. The method according to claim 2, characterized in that, The setting conditions include one or more of the following: The type of AI service is set to a specific type; The AI ​​service is a latency-sensitive AI service; To meet the latency requirements of AI services; To meet the accuracy requirements of AI services; Orchestrate transmission and / or computing resources for AI services to meet the latency requirements of AI services; The transmission and / or computing resources allocated to the AI ​​service do not meet the accuracy requirements of the AI ​​service.

4. The method according to any one of claims 1 to 3, characterized in that, The method further includes: Send the third information to the network device; or, The AI ​​service is provided to the second node based on the third information.

5. The method according to claim 1, characterized in that, The second node includes a terminal, and the method further includes: Send a first request to the second node; the first request is used to obtain the first information.

6. The method according to claim 5, characterized in that, The first request carries one or more of the following: The configuration information for reporting the first piece of information; Reporting time; The fourth information indicates that the first information is reported periodically or triggered by an event. Reporting cycle; Trigger the event.

7. The method according to claim 1, wherein the second node includes a terminal and / or an application function (AF).

8. The method according to any one of claims 1 to 3, 5 to 6, characterized in that, The first information includes one or more of the following: The first indicator, which indicates the demand for AI services; Latency of AI services; The accuracy of AI services; The first value represents the level of the AI ​​service; The first parameter is used to assist in generating QoS parameters and / or QoS policies; The first identifier indicates the importance of the first indicator.

9. The method according to claim 4, characterized in that, The third information indicates one or more of the following: The transmission resources do not meet the needs of AI services; The latency requirements for AI services are not met. Compress upstream data; Prioritize meeting the latency requirements of AI services; Prioritize meeting the accuracy requirements of AI services; Reduce the amount of data transmitted upstream; Reduce link latency; Reduce link latency without affecting accuracy; Latency requirements for AI services are not taken into account; Ensure the integrity of all data transmissions; Request alternative QoS policies.

10. A message passing method, characterized in that, Applied to the second node, the method includes: Send first information to the first node, the first information representing QoS-related information of the AI ​​service.

11. The method according to claim 10, characterized in that, The method further includes: Receive a first request sent by the first node, the first request being used to obtain the first information.

12. The method according to claim 11, characterized in that, The first request carries one or more of the following: The configuration information for reporting the first piece of information; Reporting time; The fourth information indicates that the first information is reported periodically or triggered by an event. Reporting cycle; Trigger the event.

13. The method according to claim 10, characterized in that, The second node includes a terminal and / or an AF.

14. A message transmitting device, characterized in that, include: The first receiving unit is used to receive first information sent by the second node, wherein the first information represents QoS-related information of the AI ​​service. And / or, The determining unit is configured to determine third information based on the first information and / or the second information, wherein the second information represents available transmission resources and / or time-frequency resources, and the third information represents QoS parameters and / or QoS policies.

15. A message transmitting device, characterized in that, include: The first sending unit is used to send first information to the first node, wherein the first information represents QoS-related information of the AI ​​service.

16. A first node, characterized in that, include: A first processor and a first communication interface; wherein... The first communication interface is used to receive first information sent by the second node, wherein the first information represents QoS-related information of the AI ​​service; And / or, The first processor is configured to determine third information based on the first information and / or the second information, wherein the second information represents available transmission resources and / or time-frequency resources, and the third information represents QoS parameters and / or QoS policies.

17. A second node, characterized in that, include: A second processor and a second communication interface; wherein... The second communication interface is used to send first information to the first node, wherein the first information represents QoS-related information of the AI ​​service.

18. A network device, characterized in that, This includes a processor and memory for storing computer programs that can run on the processor. When the processor is used to run the computer program, it performs the steps of the method according to any one of claims 1 to 9, or the steps of the method according to claims 10 to 13.

19. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 9, or the steps of the method according to any one of claims 10 to 13.

20. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 13.

Images