WO2022105522A1 - 一种意图维持方法及装置 - Google Patents

一种意图维持方法及装置 Download PDF

Info

Publication number
WO2022105522A1
WO2022105522A1 PCT/CN2021/125201 CN2021125201W WO2022105522A1 WO 2022105522 A1 WO2022105522 A1 WO 2022105522A1 CN 2021125201 W CN2021125201 W CN 2021125201W WO 2022105522 A1 WO2022105522 A1 WO 2022105522A1
Authority
WO
WIPO (PCT)
Prior art keywords
duration
intent
information
management module
executable command
Prior art date
Application number
PCT/CN2021/125201
Other languages
English (en)
French (fr)
Inventor
李业兴
李贤明
于益俊
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP21893670.6A priority Critical patent/EP4227827A4/en
Publication of WO2022105522A1 publication Critical patent/WO2022105522A1/zh
Priority to US18/320,104 priority patent/US20230291651A1/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0894—Policy-based network configuration management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/54—Interprogram communication
    • G06F9/546—Message passing systems or structures, e.g. queues
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/02—Knowledge representation; Symbolic representation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/02—Knowledge representation; Symbolic representation
    • G06N5/022—Knowledge engineering; Knowledge acquisition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H04L41/0813—Configuration setting characterised by the conditions triggering a change of settings
    • H04L41/0816—Configuration setting characterised by the conditions triggering a change of settings the condition being an adaptation, e.g. in response to network events
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence

Definitions

  • the embodiments of the present application relate to the field of network management, and in particular, to a method and apparatus for maintaining intent.
  • ISG ENI industry specification group experiential networked intelligence
  • ETSI European telecommunications standards institute
  • AI artificial intelligence
  • the system is empirical in that it learns from the operator's instructions to improve its actions in the future. This enables operators to automate their network configuration and monitoring processes, thereby reducing operating expenses and improving the use and maintenance of their networks.
  • An intent-based experiential networked intelligence (ENI) architecture is currently defined, as shown in Figure 1.
  • the intent system periodically determines whether the intent is achieved from the time the intent is implemented, and when the intent cannot be achieved, adjusts the executable command corresponding to the intent. However, it takes a certain period of time from the execution of the intended executable command to take effect. During this period of time, the failure to achieve the intended goal does not mean that the executable command is invalid. Therefore, there is no need to adjust the executable command. It is also meaningless to judge whether the intended goal has been achieved. Further, frequently switching the intended executable commands is not conducive to the achievement of the intended target, resulting in lower achievement efficiency of the intended target, and at the same time, it also leads to consumption of system resources.
  • Embodiments of the present application provide a method and apparatus for maintaining an intent, so as to solve the problem of low efficiency in achieving an intent target caused by a ping-pong handover executable command.
  • an embodiment of the present application provides an intent maintenance method, the method comprising:
  • the state perception module obtains the first duration corresponding to the intent; the state perception module receives the first information from the policy management module, the first information indicates that the first execution command corresponding to the intent starts to be executed; When the execution duration of the executable command reaches the first duration and the intended target is not achieved, the state perception module determines a second duration, and the second duration is greater than the first duration; The policy management module sends second information, where the second information indicates the second duration.
  • the intent-reduced system can ping-pong switch intent executable commands before the intent target is achieved, thereby improving the efficiency of the intent system in reaching the intent target and reducing resource consumption.
  • the method further includes: when the intended goal is reached before the first time period is reached, the execution time period of the first executable command is shorter than the first time period, and the state awareness module A third duration is determined, where the third duration is smaller than the first duration; the state perception module sends third information to the policy management module, where the third information indicates the third duration.
  • the intent-reduced system can ping-pong switch intent executable commands before the intent target is achieved, thereby improving the efficiency of the intent system in reaching the intent target and reducing resource consumption.
  • the third duration is determined according to the execution duration of the first executable command.
  • the state sensing module when the state sensing module determines the second duration, the state sensing module sends fourth information to the knowledge management module, where the fourth information indicates the execution duration of the first executable command The first duration is reached and the intended goal is not achieved; the state awareness module receives fifth information from the knowledge management module, the fifth information indicates the second duration.
  • the second duration can be determined through the knowledge management module.
  • the state sensing module when the state sensing module determines the third duration, the state sensing module sends sixth information to the knowledge management module, where the sixth information indicates the execution duration of the first executable command ; the state perception module receives seventh information from the knowledge management module, where the seventh information indicates the third duration.
  • the third duration can be determined through the knowledge management module.
  • the first duration is carried by eighth information, and the eighth information is used to create or implement or modify the intent; or, the first duration is configured in an intent knowledge base.
  • the first duration is determined according to the duration required to achieve the intent goal in the process of simulating the implementation of the intent.
  • an embodiment of the present application provides an intent maintenance method, the method includes: a policy management module obtains a first duration corresponding to an intent; the policy management module sends first information to a state perception module, where the first information refers to The first executable command corresponding to the schematic diagram starts to be executed; when the execution duration of the first executable command reaches the first duration and the intended target is not reached, the policy management module will The command is adjusted to the second executable command corresponding to the intent; the policy management module receives second information from the state perception module, the second information indicates the second duration, and the second duration is greater than the first duration.
  • the intent-reduced system can ping-pong switch intent executable commands before the intent target is achieved, thereby improving the efficiency of the intent system in reaching the intent target and reducing resource consumption.
  • the method further includes: when the intended goal is reached before the first time period is reached, the execution time period of the first executable command is shorter than the first time period, and the policy management module Receive third information from the state sensing module, where the third information indicates the third duration, and the third duration is smaller than the first duration.
  • the intent-reduced system can ping-pong switch intent executable commands before the intent target is achieved, thereby improving the efficiency of the intent system in reaching the intent target and reducing resource consumption.
  • the third duration is determined according to the execution duration of the first executable command.
  • the first duration is carried by eighth information, and the eighth information is used to create or implement or modify the intent; or, the first duration is configured in an intent knowledge base.
  • the first duration is determined according to the duration required to achieve the intent goal in the process of simulating the implementation of the intent.
  • an embodiment of the present application provides a method for maintaining intent.
  • the method includes: the knowledge management module receives fourth information from the state perception module, where the fourth information indicates that the execution duration of the first executable command of the schematic diagram reaches the first duration and the intended goal has not been achieved; the knowledge management module determines a second duration, the second duration is greater than the first duration; the knowledge management module sends fifth information to the state perception module, the The fifth information indicates the second duration.
  • the second duration can be determined through the knowledge management module.
  • the method further includes: the knowledge management module updates, according to the second duration, the average value of the first duration corresponding to the intent of the same type as the intent.
  • the average value of the first duration can be updated through the knowledge management module.
  • an embodiment of the present application provides an intent maintenance method, the method includes: the knowledge management module receives sixth information from the state perception module, where the sixth information refers to the execution of the first executable command when the goal of the schematic diagram is achieved duration; the knowledge management module determines a third duration according to the execution duration of the first executable command, and the third duration is less than the first duration; the knowledge management module sends seventh information to the state perception module , and the seventh information indicates the third duration.
  • the third duration can be determined through the knowledge management module.
  • the method further includes: the knowledge management module updates, according to the third duration, an average value of a first duration corresponding to an intent of the same type as the intent.
  • the average value of the first duration can be updated through the knowledge management module.
  • the present application further provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is run on a computer, the computer is made to execute the first aspect or the second aspect or the first aspect.
  • the method of the third or fourth aspect is not limited to.
  • the present application also provides a computer program product comprising a program that, when run on a computer, causes the computer to perform the method of the first or second or third or fourth aspects above.
  • the present application further provides an apparatus, comprising a processor and a memory; the memory is used for storing computer-executable instructions; the processor is used for executing the computer-executable instructions stored in the memory, so that the communication apparatus The method of the first aspect or the second aspect or the third aspect or the fourth aspect above is performed.
  • the present application further provides an apparatus, including a processor and an interface circuit; the interface circuit is configured to receive a code instruction and transmit it to the processor; the processor executes the code instruction to execute the above-mentioned first The method of the one aspect or the second aspect or the third aspect or the fourth aspect.
  • FIG. 1 is a schematic diagram of an ENI architecture applied by an embodiment of the present application.
  • FIG. 2 is an overview flowchart of an intent maintenance method provided by an embodiment of the present application
  • FIG. 4 is the second schematic flow chart of intent translation and execution provided by an embodiment of the present application.
  • FIG. 5 is one of the schematic flow charts of intent maintenance provided by an embodiment of the present application.
  • FIG. 6 is the second schematic diagram of intent maintenance flow provided by an embodiment of the present application.
  • FIG. 7 is one of schematic structural diagrams of a communication device provided by an embodiment of the present application.
  • FIG. 8 is a second schematic structural diagram of a communication device provided by an embodiment of the present application.
  • the ENI architecture includes a total of 10 functional modules, which are divided into four categories: input functional modules, output functional modules, analysis functional modules, and decision-making functional modules, as shown in Figure 1.
  • Input class function module (that is, data reading and normalization (Data Ingestion and Normalisation) module) is responsible for receiving data from external systems and normalizing data. Among them, the data reading and normalization modules are two modules.
  • Output class function module (ie, the output generation and denormalisation module) is responsible for converting the internal commands of the system into a format that can be processed by the external system and sending it to the external system.
  • Analysis function module responsible for perceiving and analyzing the existing network state and predicting the future network state.
  • the analysis function modules include the Knowledge Management module, the Context Awareness module, and the Cognition Management module.
  • the knowledge management module is mainly responsible for managing all knowledge within the scope of ENI, completing the identification of the strategy type of the intention and the storage of the knowledge related to the intention.
  • the context-aware module is mainly responsible for acquiring the status and environment information of the Assisted System, for example, for acquiring the performance data of network elements (such as base stations (gNodeB)).
  • the auxiliary system may be a radio access network element or a core network element (such as a user plane function (UPF)).
  • UPF user plane function
  • the cognitive management module is mainly responsible for the understanding of network data and the cognition of network status in the process of intention maintenance.
  • the decision-making function module is used to generate new strategies according to the intention on the basis of the perception of the network state, arrange the strategy, and send the operation command to the output function module.
  • Decision-making functional modules include Situation Awareness, Policy Management, and Model-driven Engineering.
  • the decision-making functional modules include a Situation Awareness module, a Policy Management module, a Model-driven Engineering module, and an Intent Translation module.
  • the context perception module is mainly responsible for perceiving the impact of the recommendation or command issued by the ENI system on the auxiliary system. For example, it is used to perceive the achievement of the intended goal.
  • the model-driven engineering module mainly assists in transforming intent knowledge and contextual information into a format that can be recognized by the policy management module through a model-driven approach.
  • the policy management module is mainly responsible for generating a policy to ensure the achievement of the intent based on the intent knowledge and context information, and delivering it to the Assisted System.
  • the intent translation module is mainly responsible for the lexical parsing of intent expressions, and obtains intent-related knowledge from the knowledge management module according to the content of the intent expressions.
  • the intent enforcement process includes intent translation, execution, and maintenance.
  • the intent translation module is responsible for intent translation, and sends the translation result to the policy management module, and the policy management module executes and maintains the intent.
  • the policy management module adjusts the executable command of the intent according to the achievement of the intent.
  • the expression of the intent is defined, and the expression of the intent is as follows:
  • a complete intent consists of two parts, the intent action (IntentDrivenAction) and the intent target (IntentDrivenObject).
  • the intent system periodically determines whether the intent is achieved from the time the intent is implemented, and when the intent cannot be achieved, adjusts the executable command corresponding to the intent.
  • the inspection period for the achievement of the intention goal is limited by the reporting capability of the auxiliary system, which is determined by the intention system according to the intention business, and the minimum can be up to 1min.
  • the intention system will periodically judge the intention to achieve before the intention is achieved.
  • the inspection period for the achievement of the intended goal is the same as the reporting period of the auxiliary system.
  • the reference duration involved in the embodiment of the present application refers to the minimum time interval for adjusting the intended executable command, that is, the shortest duration required for the executable command to be executed to the next time the executable command is adjusted.
  • the reference duration may also be called an expected duration (expected time), or other names, which are not limited in this embodiment of the present application.
  • the network state collection period ie the reporting period
  • an embodiment of the present application provides an intention maintenance method, as shown in FIG. 2 , the method includes: :
  • Step 200 The state perception module acquires the first duration corresponding to the intent.
  • the first duration may be carried by a message for creating an intent or implementing an intent or modifying an intent.
  • the intent creator can send an intent message to the intent system, the message includes the first duration, and further, the first duration can be forwarded to the state awareness module by other modules in the intent system.
  • the first duration is the initial reference duration.
  • the first duration can also be configured in the intent knowledge base.
  • other modules in the intent system can obtain the first duration from the intent knowledge base and forward it to the state-aware module.
  • the first duration may refer to the average duration of the same type of intention adjustment executable command.
  • the first duration may also be determined according to the duration required to achieve the intent goal during the implementation of the simulated intent.
  • the first duration at this time is the initial reference duration.
  • the reference duration is adjusted, that is, the reference duration is reduced.
  • adjust the reference duration that is, extend the reference duration.
  • a reference duration may be determined as the first duration. For example, assuming that the preset floating range is 40s, after several times of simulating the intent implementation process, the variation range of the reference duration is between 4 minutes and 45 seconds to 5 minutes and 15 seconds. At this time, the floating range of the reference duration is 30s, which falls within the preset range. Floating range, you can set the first duration to 5 minutes.
  • the last obtained reference duration is used as the first duration. For example, assuming that the preset number of times is 5, when the simulation intention implementation process reaches 5 times, the reference duration determined after the 5th simulation intention implementation process is used as the first duration.
  • Step 210 The policy management module sends first information to the state perception module, where the first information indicates that the first execution command corresponding to the schematic diagram starts to be executed.
  • the state sensing module monitors the execution duration of the first executable command. If the execution duration of the first executable command reaches the first duration and the intended target is not achieved, step 220A is executed. If the intended goal is achieved before the first time period arrives, step 220B is executed. Exemplarily, after receiving the first information, the state perception module starts to count the execution duration of the first executable command.
  • Step 220A When the execution duration of the first executable command reaches the first duration and the intended target is not achieved, the state perception module determines a second duration, which is greater than the first duration.
  • the state awareness module may itself determine the second duration.
  • the state awareness module may send fourth information to the knowledge management module, the fourth information indicates that the execution duration of the first executable command has reached the first duration and the intended target has not been achieved.
  • the knowledge management module determines the second duration after receiving the fourth information, and sends fifth information to the state perception module, where the fifth information includes the second duration.
  • the state awareness module does not need to send the fourth information to the knowledge management module, but when the intended goal is reached before the first time period arrives, the state awareness module sends third information to the policy management module, the third information indicating the first
  • the execution time of an executable command is the time required to achieve the actual intended goal.
  • the knowledge management module determines the second duration when determining that the third information is not received, and sends fifth information to the state management module, where the fifth information includes the second duration.
  • the knowledge management module can also adjust the average duration of the executable command according to the second duration update intention of the same type.
  • state perception module or the knowledge management module can use the following formula to determine the second duration:
  • Time_update Time_initial+ ⁇
  • ⁇ can be a preset value, and ⁇ is greater than 0.
  • Time_update represents the second duration
  • state perception module or the knowledge management module may use other formulas to determine the second duration, which is not limited in this embodiment of the present application.
  • Step 230A The state perception module sends second information to the policy management module, where the second information indicates the second duration.
  • Step 240A When the execution duration of the first executable command reaches the first duration and the intended target is not achieved, the policy management module adjusts the first executable command to the second executable command.
  • step 230A and step 240A does not limit the execution order of step 230A and step 240A.
  • the policy management module needs to obtain the first duration corresponding to the intent first, and the specific manner of acquiring the first duration by the policy management module may refer to step 200, and the repetition will not be repeated.
  • the policy management module determines whether the first executable command needs to be adjusted according to whether the intended target is achieved within the first time period. If the execution duration of the first executable command reaches the first duration and the intended target is not achieved, that is, the intended target is not achieved within the first duration, the policy management module needs to adjust the executable command, that is, the intended first executable command Adjusted to the second executable command corresponding to the intent. It should be noted that one intent may correspond to multiple executable commands, and the multiple executable commands here are a first executable command and a second executable command. If the intended goal is achieved before the first time period is reached, the policy management module does not need to adjust the executable command.
  • the policy management module sends a message for instructing the second execution command to start execution to the state management module.
  • the policy management module determines whether the second executable command needs to be adjusted according to whether the intended target is achieved within the second time period, and enters a new loop process.
  • the following only takes the switching executable command as an example for description, and the embodiments of the present application are also applicable to switching rules or switching sub-intents.
  • the policy management module adjusts the first rule to the second rule.
  • the policy management module adjusts the first sub-intent to the second sub-intent.
  • Step 220B When the intended target is reached before the first duration, the execution duration of the first executable command is shorter than the first duration, and the state perception module determines a third duration, which is shorter than the first duration.
  • the state sensing module can determine the third duration by itself.
  • the state perception module may send sixth information to the knowledge management module, where the sixth information indicates the execution duration of the first executable command.
  • the state perception module receives seventh information from the knowledge management module, where the seventh information indicates the third duration.
  • the knowledge management module can also adjust the average duration of the executable command according to the third duration update intention of the same type.
  • the third duration is the execution duration of the first executable command.
  • the state perception module or the knowledge management module may use the following formula to determine the third duration:
  • state perception module or the knowledge management module may use other formulas to determine the third duration, which is not limited in this embodiment of the present application.
  • Step 230B The state perception module sends third information to the policy management module, where the third information indicates a third duration.
  • the policy management module determines whether the first executable command needs to be adjusted according to whether the intended target is achieved within the third time period.
  • step 220B and step 230B are optional steps. If the intended target is reached before the first duration, the state perception module may not need to determine the third duration, and the policy management module continues to determine whether to adjust the first executable command according to the first duration.
  • the intent-reduced system can ping-pong switch intent executable commands before the intent target is achieved, thereby improving the efficiency of the intent system in reaching the intent target and reducing resource consumption.
  • FIG. 3 is one of the schematic flow charts of intent translation and execution.
  • the initial reference duration is issued by the intent creator.
  • Step 1 The intent creator sends a new policy (newPolicy) message to the data reading and normalization module, and the new policy message includes the initial reference duration.
  • the initial reference duration may refer to the first duration in the embodiment shown in FIG. 2 .
  • the initial reference duration may also be referred to as an initial value of the reference duration.
  • the intent creator may be a user or an application or an operations support system (OSS) or a business support system (BSS).
  • OSS operations support system
  • BSS business support system
  • Step 2 The data reading and normalization module normalizes the new policy message and related data, and sends the normalized data to the knowledge management module.
  • the normalized data also includes the initial reference duration.
  • Step 3 The knowledge management module updates the intent knowledge base according to the normalized data.
  • Step 4 The knowledge management module sends a translation policy (translatePolicy) request message to the intent translation module.
  • the translation request message includes the initial reference duration.
  • Step 5 The intent translation module assigns an identifier to the intent policy (Intent Policy).
  • Step 6 The intent translation module sends an identification update request to the knowledge management module, and the identification update request is used to update the intent identification in the intent knowledge base.
  • Step 7 The intent translation module sends a response message to the output generation and denormalization module, and the response message is used to inform that the intent has started to be processed.
  • Step 8 The output generation and denormalization module sends a response message to the intent creator, which is used to inform that the intent has started to be processed.
  • the intent translation module also needs to perform intent syntax and lexical checking. If there is an error in the grammar and lexical check, the intent translation module sends error feedback to the intent creator. The intent translation module sends an intent error message to the knowledge management module. If the intent syntax lexical check is correct, proceed to step 9.
  • Step 9 The intent translation module extracts intent keywords.
  • Step 10 The intent translation module sends an intent keyword update message to the knowledge management module.
  • Step 11 The intent translation module sends an intent knowledge acquisition request to the knowledge management module.
  • Step 12 The knowledge management module sends the intent knowledge to the intent translation module.
  • Step 13 The intent translation module arranges the intent knowledge to determine the intent translation result.
  • the intent knowledge may include intent target information, names of network state parameters, identifiers of network elements related to intent policies, and the like.
  • Step 14 The intent translation module sends an update intent policy instance context (updPolsemantics) message to the knowledge management module.
  • update intent policy instance context updPolsemantics
  • Step 15 The intent translation module sends an intent policy translation complete (polComplete) message to the context awareness module.
  • the intent-policy translation complete (polComplete) message includes the initial reference duration and the intent translation result.
  • Step 16 The context sensing module judges the network state, and sends a network state notification completion (ctxComplete) message to the state sensing module.
  • the network status notification complete (ctxComplete) message includes the initial reference duration and the intent translation result.
  • Step 17 The state perception module judges whether the intended goal is achieved, and sends a network state notification completion message to the model-driven engineering module.
  • the network status notification complete (ctxComplete) message includes the initial reference duration and the intent translation result.
  • the state perception module obtains the initial reference duration to prepare for the step 20 .
  • Step 18 The model-driven engineering module converts the intent translation result into a form that can be understood by the intent-acting entity (ie, the auxiliary system) according to the network status notification completion message, updates it to the knowledge management module, and sends the schema to the policy management module Drive project complete (mdeComplete) message.
  • the mode-driven project completion (mdeComplete) message includes the initial reference duration and the intent translation result.
  • Step 19 The policy management module converts and makes a decision on the intent translation result, and sends the obtained executable command to the output generation and denormalization module.
  • the executable command here may correspond to the first executable command in the implementation shown in FIG. 2 above.
  • Step 20 The policy management module sends an intention execution start (exeBegin) message to the state perception module, and the intention execution start message indicates that the executable command in the schematic diagram starts to be executed.
  • intention execution start (exeBegin) message
  • Step 21 The state awareness module starts timing according to the received intent execution start message.
  • the state awareness module records the execution duration of the executable command. If the intended target is not achieved within the reference duration, the state awareness module may record that the execution duration of the executable command reaches the reference duration and the intended target is not achieved, or make no record.
  • Step 22 The output generation and denormalization module sends the executable command to the intended entity (ie, the auxiliary system).
  • the reference duration can be carried when the intent is first issued, which can reduce the probability that the intent can adjust the intent to execute the command before reaching the goal, and reduce the consumption of resources.
  • FIG. 4 is a second schematic flow chart of intent translation and execution.
  • the initial reference duration is stored in the intent knowledge base.
  • Step 1 The intent creator sends a new policy (newPolicy) message to the data reading and normalization module, the new policy message including the reference duration.
  • newPolicy new policy
  • the intent creator may be a user or an application or an OSS or a BSS.
  • Step 2 The data reading and normalization module normalizes the new policy message and related data, and sends the normalized data to the knowledge management module. Among them, the normalized data also includes the reference duration.
  • Step 3 The knowledge management module updates the intent knowledge base according to the normalized data.
  • Step 4 The knowledge management module sends a translation policy (translatePolicy) request message to the intent translation module.
  • the translation request message includes the reference duration.
  • Step 5 The intent translation module assigns an identifier to the intent policy (Intent Policy).
  • Step 6 The intent translation module sends an identification update request to the knowledge management module, and the identification update request is used to update the intent identification in the intent knowledge base.
  • Step 7 The intent translation module sends a response message to the output generation and denormalization module, and the response message is used to inform that the intent has started to be processed.
  • Step 8 The output generation and denormalization module sends a response message to the intent creator, which is used to inform that the intent has started to be processed.
  • the intent translation module also needs to perform intent syntax and lexical checking. If there is an error in the grammar and lexical check, the intent translation module sends error feedback to the intent creator. The intent translation module sends an intent error message to the knowledge management module. Also, if the intent syntax is lexically checked, proceed to step 9.
  • Step 9 The intent translation module extracts intent keywords.
  • Step 10 The intent translation module sends an intent keyword update message to the knowledge management module.
  • Step 11 The intent translation module sends an intent knowledge acquisition request to the knowledge management module.
  • Step 12 The knowledge management module sends a return intent (returnIntent) message to the intent translation module, where the message carries the initial reference duration and intent knowledge.
  • returnIntent return intent
  • the initial reference duration is stored in the intent knowledge base.
  • the initial reference duration may be the reference duration corresponding to the type of intent, that is, the reference duration of a type of intent.
  • Step 13 The intent translation module arranges the intent knowledge to determine the intent translation result.
  • Intent knowledge may include intent target information, names of network state parameters and identifiers of network elements related to intent policies, etc.
  • Step 14 The intent translation module sends an update intent policy instance context (updPolsemantics) message to the knowledge management module.
  • update intent policy instance context updPolsemantics
  • Step 15 The intent translation module sends an intent policy translation complete (polComplete) message to the context awareness module.
  • the intent-policy translation complete (polComplete) message includes the initial reference duration and the intent translation result.
  • Step 16 The context sensing module judges the network state, and sends a network state notification completion (ctxComplete) message to the state sensing module.
  • the network status notification complete (ctxComplete) message includes the initial reference duration and the intent translation result.
  • Step 17 The state perception module judges whether the intended goal is achieved, and sends a network state notification completion message to the model-driven engineering module.
  • the network status notification complete (ctxComplete) message includes the initial reference duration and the intent translation result.
  • Step 18 The model-driven engineering module converts the intent translation result into a form that can be understood by the intent-acting entity (ie, the auxiliary system) according to the network status notification completion message, updates it to the knowledge management module, and sends the schema to the policy management module Drive project complete (mdeComplete) message.
  • the mode-driven project completion (mdeComplete) message includes the initial reference duration and the intent translation result.
  • Step 19 The policy management module converts and makes a decision on the intent translation result, and the obtained executable command or rule is sent to the output generation and denormalization module.
  • Step 20 The policy management module sends an intention execution start (exeBegin) message to the state perception module, and the intention execution start message indicates that the executable command in the schematic diagram starts to be executed.
  • intention execution start (exeBegin) message
  • Step 21 The state awareness module starts timing according to the received intent execution start message.
  • the state awareness module records the execution duration of the executable command. If the intended target is not achieved within the reference duration, the state awareness module records that the execution duration of the executable command reaches the reference duration and the intended target is not achieved.
  • Step 22 The output generation and denormalization module sends executable commands or rules to the intended entity (ie, the auxiliary system).
  • the reference duration is stored in the intent knowledge base, which can reduce the probability that the intent can adjust the intent to execute the command before reaching the goal, and reduce the consumption of resources.
  • Embodiment 3 As shown in FIG. 5 , one of the flow charts of intent maintenance is shown.
  • the knowledge management module can determine the updated reference duration.
  • Step 1 The context-aware module sends a performance data subscription request to the output generation and denormalization module.
  • Step 2 The output generation and denormalization module sends a performance data subscription request to the auxiliary system.
  • Step 3 The auxiliary system sends performance data to the data reading and normalization module.
  • Step 4 The data reading and normalization module normalizes the performance data, and sends the processed data to the knowledge management module.
  • Step 5 The knowledge management module sends the policy and performance data (polAndPerformanceData) to the context awareness module, and the policy and performance data include the updated reference duration.
  • policy and performance data include the updated reference duration.
  • the state perception module can send an update policy scenario (updPolSA) message to the knowledge management module, and the message includes the measurement duration.
  • updPolSA update policy scenario
  • the knowledge management module determines the updated reference duration according to the measurement duration. It can be understood that the measurement duration here may correspond to the execution duration of the first executable command in the embodiment shown in FIG. 2 . At this time, the updated reference duration may correspond to the third duration in the embodiment shown in FIG. 2 .
  • the calculation formula of the updated reference duration is:
  • Time_update Time_initial+w*t, where Time_update is the updated reference duration, and Time_initial is the reference duration recorded in the state perception module, that is, the latest updated reference duration or the initial reference duration.
  • the updated reference duration can be directly defined to be equal to the measurement duration.
  • This embodiment of the present application does not limit the timing at which the state perception module sends the measurement duration to the knowledge management module.
  • the status awareness module receives the network status notification completion message from the context awareness module, that is, after step 6, the status awareness module sends an update policy context message to the knowledge management module, where the message includes the measurement duration.
  • the measurement duration at this time is the execution duration of the previous round of executable command execution.
  • the state awareness module can send updPolSA to the knowledge management module, which indicates that the reference duration is reached and the intended target is not achieved, and the knowledge management module determines the updated reference duration according to the message.
  • Time_update Time_initial+ ⁇
  • Time_update is the updated reference duration
  • Time_initial is the initial reference duration
  • ⁇ may be a preset value
  • ⁇ is greater than 0.
  • the updated reference duration may correspond to the second duration in the embodiment shown in FIG. 2 .
  • the state awareness module may not send a message to the knowledge management module when the reference duration is reached and the intended goal is not achieved. If the knowledge management module does not receive the message sent by the state perception module, it determines that the reference duration has been reached and the intended target has not been achieved, and the knowledge management module determines the updated reference duration.
  • the knowledge management module may also update the initial reference duration of the intent belonging to the same category in the intent system according to the updated reference duration.
  • T (w1*T1+w2*T2+w3*T3+...+wn*Tn)/n, where n is an intent in the intent system that belongs to the same class (eg, increase bandwidth class intent) and is active
  • the number of intents, T1, T2, ... Tn, etc. are the initial reference durations corresponding to the n intents respectively.
  • w1, w2...wn are the weights corresponding to T1, T2,...Tn respectively.
  • Step 6 The context sensing module sends a network state notification completion message to the state sensing module, and the network state notification completion message includes the updated reference duration.
  • Step 7 The state perception module sends a network state notification completion message to the model-driven engineering module, and the network state notification completion message includes the updated reference duration.
  • Step 8 The model-driven engineering module sends a mode-driven engineering completion message to the policy management module, and the mode-driven engineering completion message includes the updated reference duration.
  • Step 9 The policy management module determines whether to adjust the intended executable command according to the updated reference duration and the achievement of the intended target. If the updated reference duration is not reached and the intended target is achieved, there is no need to adjust the executable command of the intent, and the process ends. If the updated reference duration is reached and the intended target is not achieved, the executable command of the intent needs to be adjusted, and step 10 is executed.
  • Step 10 The policy management module sends the adjusted executable command to the output generation and denormalization module.
  • Step 11 The output generation and denormalization module sends the adjusted executable command to the auxiliary system.
  • Step 12 The policy management module sends an intention execution start message to the state sensing module, notifying the state sensing module to start a new round of execution time timing of the executable command.
  • the knowledge management module can update the reference duration.
  • Embodiment 4 As shown in FIG. 6 , the second schematic diagram of the intent maintenance process is shown.
  • the state sensing module may determine the updated reference duration.
  • Step 1 The context-aware module sends a performance data subscription request to the output generation and denormalization module.
  • Step 2 The output generation and denormalization module sends a performance data subscription request to the auxiliary system.
  • Step 3 The auxiliary system sends performance data to the data reading and normalization module.
  • Step 4 The data reading and normalization module normalizes the performance data, and sends the processed data to the knowledge management module.
  • Step 5 The knowledge management module stores the performance data, and sends the performance data and intent context information to the context awareness module, wherein the intent context information is generated in the intent translation and execution stages.
  • Step 6 The context awareness module judges the network state according to the performance data, adds the network state information to the context information, updates the intent context information to the knowledge management module, and sends the updated context information to the state awareness module.
  • Step 7 The state perception module determines the updated reference duration.
  • the state perception module obtains the intended target achievement time according to the actual measurement, that is, when the intended target is achieved within the reference duration, the execution duration of the executable command (hereinafter referred to as the measurement duration), Calculate the updated reference duration.
  • the calculation formula of the updated reference duration is:
  • the updated reference duration can be defined to be equal to the measurement duration.
  • Time_update Time_initial+ ⁇
  • Time_update is the updated reference duration
  • Time_initial is the reference duration recorded in the state perception module, that is, the last updated reference duration or the initial reference duration
  • ⁇ can be a preset value, ⁇ Greater than 0.
  • Step 8 The state perception module sends a network state notification completion message to the model-driven engineering module, and the network state notification completion message includes the updated reference duration.
  • Step 9 The model-driven engineering module sends a mode-driven engineering completion message to the policy management module, and the driving engineering completion message includes the updated reference duration.
  • Step 10 The policy management module determines whether to adjust the intended executable command according to the updated reference duration and the achievement of the intended target. If the updated reference duration is not reached and the intended target is achieved, there is no need to adjust the executable command of the intent, and the process ends. If the updated reference duration is reached and the intended target is not achieved, the executable command of the intent needs to be adjusted, and step 11 is executed.
  • Step 11 The policy management module sends the adjusted executable command to the output generation and denormalization module.
  • Step 12 The output generation and denormalization module sends the adjusted executable command to the auxiliary system.
  • Step 13 The policy management module sends an intention execution start message to the state sensing module, notifying the state sensing module to start a new round of execution time timing of executable commands.
  • the knowledge management module can update the reference duration.
  • the network device and the terminal device include corresponding hardware structures and/or software modules for performing each function.
  • the units and method steps of each example described in conjunction with the embodiments disclosed in the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is performed by hardware or computer software-driven hardware depends on the specific application scenarios and design constraints of the technical solution.
  • FIG. 7 and FIG. 8 are schematic structural diagrams of possible communication apparatuses provided by embodiments of the present application. These communication apparatuses can be used to implement the functions of the state perception module, the policy management module, or the knowledge management module in the above method embodiments, so the beneficial effects of the above method embodiments can also be achieved.
  • the communication device may be a state perception module, a policy management module or a knowledge management module as shown in FIG. 1 , and may also be a chip applied to a state perception module, a policy management module or a knowledge management module.
  • the communication apparatus 700 includes a processing unit 710 and a transceiver unit 720 .
  • the communication apparatus 700 is used to implement the function of the state sensing module in the method embodiment shown in the above-mentioned FIG. 2 , FIG. 3 , FIG. 4 , FIG. 5 or FIG.
  • the processing unit 710 is configured to, when the execution duration of the first executable command reaches the first duration and the intended target is not achieved, the state perception module determines a second duration, which is greater than the first duration.
  • the transceiver unit 720 is configured to send second information to the policy management module, where the second information indicates the second duration.
  • the transceiver unit 720 is used to obtain the first duration corresponding to the intent, and send the status to the state.
  • the sensing module sends first information, where the first information indicates that the first executable command corresponding to the schematic diagram starts to be executed;
  • the processing unit 710 is configured to adjust the first executable command to a second executable command corresponding to the intent when the execution duration of the first executable command reaches the first duration and the intended target is not reached;
  • the transceiver unit 720 is configured to receive second information from the state sensing module, where the second information indicates a second duration, and the second duration is greater than the first duration.
  • the transceiver unit 720 is configured to receive the fourth information from the state perception module , the fourth information means that the execution duration of the first executable command of the schematic diagram reaches the first duration and the intended target has not been achieved;
  • the processing unit 710 is configured to determine a second duration, where the second duration is greater than the first duration
  • the transceiver unit 720 is configured to send fifth information to the state sensing module, where the fifth information indicates the second duration.
  • the transceiver unit 720 is configured to receive the sixth information from the state perception module , the sixth information refers to the execution duration of the first executable command when the goal of the schematic diagram is achieved;
  • the processing unit 710 is configured to determine a third duration according to the execution duration of the first executable command, where the third duration is smaller than the first duration;
  • the transceiver unit 720 is configured to send seventh information to the state sensing module, where the seventh information indicates the third duration.
  • processing unit 710 and the transceiver unit 720 can be obtained directly by referring to the related descriptions in the method embodiments shown in FIG. 2 , FIG. 3 , FIG. 4 , FIG. 5 or FIG. 6 , and details are not repeated here.
  • the communication apparatus 800 includes a processor 810 and an interface circuit 820 .
  • the processor 810 and the interface circuit 820 are coupled to each other.
  • the interface circuit 820 can be a transceiver or an input-output interface.
  • the communication apparatus 800 may further include a memory 830 for storing instructions executed by the processor 810 or input data required by the processor 810 to execute the instructions or data generated after the processor 810 executes the instructions.
  • the processor 810 is used to implement the function of the above-mentioned processing unit 710
  • the interface circuit 820 is used to implement the above-mentioned transceiver unit 720 function.
  • the processor in the embodiments of the present application may be a central processing unit (Central Processing Unit, CPU), and may also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application-specific integrated circuits (Application Specific Integrated Circuit, ASIC), Field Programmable Gate Array (Field Programmable Gate Array, FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof.
  • a general-purpose processor may be a microprocessor or any conventional processor.
  • the method steps in the embodiments of the present application may be implemented in a hardware manner, or may be implemented in a manner in which a processor executes software instructions.
  • Software instructions can be composed of corresponding software modules, and software modules can be stored in random access memory (Random Access Memory, RAM), flash memory, read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM) , PROM), Erasable Programmable Read-Only Memory (Erasable PROM, EPROM), Electrically Erasable Programmable Read-Only Memory (Electrically Erasable Programmable Read-Only Memory (Electrically EPROM, EEPROM), registers, hard disks, removable hard disks, CD-ROMs or known in the art in any other form of storage medium.
  • RAM Random Access Memory
  • ROM read-only memory
  • PROM programmable read-only memory
  • PROM Erasable Programmable Read-Only Memory
  • EPROM Electrically Erasable Programmable Read-Only Memory
  • An exemplary storage medium is coupled to the processor, such that the processor can read information from, and write information to, the storage medium.
  • the storage medium can also be an integral part of the processor.
  • the processor and storage medium may reside in an ASIC.
  • the ASIC may be located in a network device or in an end device.
  • the processor and the storage medium may also exist in the network device or the terminal device as discrete components.
  • the above-mentioned embodiments it may be implemented in whole or in part by software, hardware, firmware or any combination thereof.
  • software it can be implemented in whole or in part in the form of a computer program product.
  • the computer program product includes one or more computer programs or instructions.
  • the processes or functions described in the embodiments of the present application are executed in whole or in part.
  • the computer may be a general purpose computer, a special purpose computer, a computer network, network equipment, user equipment, or other programmable apparatus.
  • the computer program or instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program or instructions may be downloaded from a website site, computer, A server or data center transmits by wire or wireless to another website site, computer, server or data center.
  • the computer-readable storage medium may be any available medium that can be accessed by a computer, or a data storage device such as a server, data center, or the like that integrates one or more available media.
  • the usable medium can be a magnetic medium, such as a floppy disk, a hard disk, and a magnetic tape; it can also be an optical medium, such as a digital video disc (DVD); it can also be a semiconductor medium, such as a solid state drive (solid state drive). , SSD).
  • a magnetic medium such as a floppy disk, a hard disk, and a magnetic tape
  • an optical medium such as a digital video disc (DVD)
  • DVD digital video disc
  • it can also be a semiconductor medium, such as a solid state drive (solid state drive). , SSD).
  • “at least one” means one or more, and “plurality” means two or more.
  • “And/or”, which describes the relationship of the associated objects, indicates that there can be three kinds of relationships, for example, A and/or B, it can indicate that A exists alone, A and B exist at the same time, and B exists alone, where A, B can be singular or plural.
  • the character “/” generally indicates that the related objects are a kind of "or” relationship; in the formula of this application, the character "/” indicates that the related objects are a kind of "division” Relationship.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Telephonic Communication Services (AREA)
  • Stored Programmes (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

一种意图维持方法及装置，该方法包括：状态感知模块获取意图对应的第一时长。状态感知模块接收来自于策略管理模块的第一信息，第一信息指示意图对应的第一执行命令开始执行。在第一可执行命令的执行时长到达第一时长且意图的目标未达成时，状态感知模块确定第二时长，第二时长大于第一时长；状态感知模块向策略管理模块发送第二信息，第二信息指示第二时长。采用上述方法可以实现减少意图系统在意图目标达成前乒乓切换意图可执行命令，提高意图系统达成意图目标的效率，减少资源的消耗。

Description

一种意图维持方法及装置
相关申请的交叉引用
本申请要求在2020年11月18日提交中国专利局、申请号为202011295591.1、申请名称为“一种意图维持方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请实施例涉及网络管理领域，尤其涉及一种意图维持方法及装置。
背景技术
欧洲电信标准协会(european telecommunications standards institute，ETSI)成立的体验式网络化智能行业规范组(industry specification group experiential networked intelligence，ISG ENI)工作组的目标是定义一个基于观察方向决定行为控制模型的认知网络管理体系架构。它使用人工智能(artificial intelligence，AI)技术和情境感知策略，根据用户需求、环境条件和业务目标的变化调整提供的服务。该系统是经验性的，因为它从操作者的指令中学习相关知识，以改进其在将来采取的行动。因而能够帮助运营商自动化其网络配置和监控过程，从而减少运营支出并改善其网络的使用和维护。当前定义了基于意图(intent)的体验式网络智能(experiential networked intelligence，ENI)架构，如图1所示。
意图系统从意图实施开始周期性地判断意图是否达成，当意图不能够达成时，调整意图对应的可执行命令。但是，意图的可执行命令从开始执行到生效需要一定的时间，在此段时间内，意图的目标未达成并不是可执行命令无效，因此，不需要调整可执行命令，且在此段时间内判断意图的目标是否达成也是没有意义的。进一步地，频繁地切换意图的可执行命令不利于意图的目标的达成，导致意图的目标的达成效率较低，同时，还会导致系统资源的消耗。
发明内容
本申请实施例提供一种意图维持方法及装置，用以解决乒乓切换可执行命令导致意图的目标的达成效率较低的问题。
第一方面，本申请实施例提供一种意图维持方法，该方法包括：
状态感知模块获取意图对应的第一时长；所述状态感知模块接收来自于策略管理模块的第一信息，所述第一信息指示所述意图对应的第一执行命令开始执行；在所述第一可执行命令的执行时长到达所述第一时长且所述意图的目标未达成时，所述状态感知模块确定第二时长，所述第二时长大于所述第一时长；所述状态感知模块向所述策略管理模块发送第二信息，所述第二信息指示所述第二时长。
采用上述方法可以实现减少意图系统在意图目标达成前乒乓切换意图可执行命令，提高意图系统达成意图目标的效率，减少资源的消耗。
在一种可能的设计中，还包括：在所述意图的目标在所述第一时长到达前达成时，所 述第一可执行命令的执行时长小于所述第一时长，所述状态感知模块确定第三时长，所述第三时长小于所述第一时长；所述状态感知模块向所述策略管理模块发送第三信息，所述第三信息指示所述第三时长。
采用上述方法可以实现减少意图系统在意图目标达成前乒乓切换意图可执行命令，提高意图系统达成意图目标的效率，减少资源的消耗。
在一种可能的设计中，所述第三时长是根据所述第一可执行命令的执行时长确定的。
在一种可能的设计中，在所述状态感知模块确定第二时长时，所述状态感知模块向知识管理模块发送第四信息，所述第四信息指示所述第一可执行命令的执行时长到达所述第一时长且所述意图的目标未达成；所述状态感知模块接收来自于所述知识管理模块的第五信息，所述第五信息指示所述第二时长。
采用上述方法可以实现通过知识管理模块确定第二时长。
在一种可能的设计中，在所述状态感知模块确定第三时长时，所述状态感知模块向知识管理模块发送第六信息，所述第六信息指示所述第一可执行命令的执行时长；所述状态感知模块接收来自于所述知识管理模块的第七信息，所述第七信息指示所述第三时长。
采用上述方法可以实现通过知识管理模块确定第三时长。
在一种可能的设计中，所述第一时长由第八信息携带，所述第八信息用于创建或实施或修改所述意图；或者，所述第一时长配置在意图知识库中。
在一种可能的设计中，所述第一时长是根据模拟所述意图实施过程中意图目标达成所需的时长确定的。
第二方面，本申请实施例提供一种意图维持方法，该方法包括：策略管理模块获取意图对应的第一时长；所述策略管理模块向状态感知模块发送第一信息，所述第一信息指示意图对应的第一可执行命令开始执行；在所述第一可执行命令的执行时长到达所述第一时长且所述意图的目标未达成时，所述策略管理模块将所述第一可执行命令调整为所述意图对应的第二可执行命令；所述策略管理模块接收来自于所述状态感知模块的第二信息，所述第二信息指示所述第二时长，所述第二时长大于所述第一时长。
采用上述方法可以实现减少意图系统在意图目标达成前乒乓切换意图可执行命令，提高意图系统达成意图目标的效率，减少资源的消耗。
在一种可能的设计中，还包括：在所述意图的目标在所述第一时长到达前达成时，所述第一可执行命令的执行时长小于所述第一时长，所述策略管理模块接收来自于所述状态感知模块的第三信息，所述第三信息指示所述第三时长，所述第三时长小于所述第一时长。
采用上述方法可以实现减少意图系统在意图目标达成前乒乓切换意图可执行命令，提高意图系统达成意图目标的效率，减少资源的消耗。
在一种可能的设计中，所述第三时长是根据所述第一可执行命令的执行时长确定的。
在一种可能的设计中，所述第一时长由第八信息携带，所述第八信息用于创建或实施或修改所述意图；或者，所述第一时长配置在意图知识库中。
在一种可能的设计中，所述第一时长是根据模拟所述意图实施过程中意图目标达成所需的时长确定的。
第三方面，本申请实施例提供一种意图维持方法，该方法包括：知识管理模块接收来自于状态感知模块的第四信息，第四信息指示意图的第一可执行命令的执行时长到达第一时长且所述意图的目标未达成；所述知识管理模块确定第二时长，所述第二时长大于所述 第一时长；所述知识管理模块向所述状态感知模块发送第五信息，所述第五信息指示所述第二时长。
采用上述方法可以实现通过知识管理模块确定第二时长。
在一种可能的设计中，还包括：所述知识管理模块根据所述第二时长更新与所述意图的类型相同的意图所对应的第一时长平均值。
采用上述方法可以实现通过知识管理模块更新第一时长的平均值。
第四方面，本申请实施例提供一种意图维持方法，该方法包括：知识管理模块接收来自于状态感知模块的第六信息，第六信息指示意图的目标被达成时第一可执行命令的执行时长；所述知识管理模块根据所述第一可执行命令的执行时长确定第三时长，所述第三时长小于所述第一时长；所述知识管理模块向所述状态感知模块发送第七信息，所述第七信息指示所述第三时长。
采用上述方法可以实现通过知识管理模块确定第三时长。
在一种可能的设计中，还包括：所述知识管理模块根据所述第三时长更新与所述意图的类型相同的意图所对应的第一时长平均值。
采用上述方法可以实现通过知识管理模块更新第一时长的平均值。
第五方面，本申请还提供一种计算机可读存储介质，该计算机可读存储介质存储有计算机程序，当该计算机程序在计算机上运行时，使得计算机执行上述第一方面或第二方面或第三方面或第四方面的方法。
第六方面，本申请还提供一种包含程序的计算机程序产品，当其在计算机上运行时，使得计算机执行上述第一方面或第二方面或第三方面或第四方面的方法。
第七方面，本申请还提供一种装置，包括处理器和存储器；所述存储器用于存储计算机执行指令；所述处理器用于执行所述存储器所存储的计算机执行指令，以使所述通信装置执行上述第一方面或第二方面或第三方面或第四方面的方法。
第八方面，本申请还提供一种装置，包括处理器和接口电路；所述接口电路，用于接收代码指令并传输至所述处理器；所述处理器运行所述代码指令以执行上述第一方面或第二方面或第三方面或第四方面的方法。
附图说明
图1为本申请的实施例应用的ENI架构示意图；
图2为本申请实施例提供的意图维持方法的概述流程图；
图3为本申请实施例提供的意图转译及执行的流程示意图之一；
图4为本申请实施例提供的意图转译及执行的流程示意图之二；
图5为本申请实施例提供的意图维持流程示意图之一；
图6为本申请实施例提供的意图维持流程示意图之二；
图7为本申请实施例提供的通信装置结构示意图之一；
图8为本申请实施例提供的通信装置结构示意图之二。
具体实施方式
下面首先对本申请实施例涉及的ENI架构进行简要介绍。
ENI架构共包含10个功能模块，分为输入类功能模块、输出类功能模块、分析类功能模块以及决策类功能模块四大类，如图1所示。
输入类功能模块：(即数据读取和规范化(Data Ingestion and Normalisation)模块)负责接收外部系统的数据以及对数据进行归一化等操作。其中，数据读取和规范化模块是两个模块。
输出类功能模块：(即输出生成和非规范化(Output Generation and Denormalisation)模块)负责将系统内部命令转化为外部系统能够处理的格式并发送给外部系统。
分析类功能模块：负责对现有的网络状态进行感知分析并且预测未来的网络状态。分析类功能模块包括，知识管理(Knowledge Management)模块、上下文感知模块(Context Awareness)模块、认知管理(Cognition Management)模块。
其中，知识管理模块主要负责管理ENI范围内的所有知识，完成意图的策略类型的识别和意图相关知识的存储。
上下文感知模块主要负责获取辅助系统(Assisted System)的状态和环境信息，例如，用于获取网元(如基站(gNodeB))的性能数据。其中，辅助系统可以是无线接入网元或者核心网网元(如用户面功能(user plane function,UPF))。
认知管理模块主要负责意图维持过程中网络数据的理解和网络状态认知。
决策类功能模块用于在对网络状态感知的基础上，根据意图生成新的策略并对策略进行编排，并将操作命令发送给输出类功能模块。决策类功能模块包括情景感知(Situation Awareness)模块、策略管理(Policy Management)模块、模型驱动工程(Model-driven Engineering)模块。或者，决策类功能模块包括情景感知(Situation Awareness)模块、策略管理(Policy Management)模块、模型驱动工程(Model-driven Engineering)模块、意图转译(Intent Translation)模块。
情景感知模块主要负责感知ENI系统下发的推荐或命令对辅助系统的影响。例如，用于感知意图目标的达成情况。
模型驱动工程模块主要辅助将意图知识和上下文信息通过模型驱动方法转化为策略管理模块能够识别的格式。
策略管理模块主要负责根据意图知识和上下文信息，产生保证意图达成的策略，并向Assisted System下发。
意图转译模块主要负责意图表达式的词法语法解析，并根据意图表达式的内容，从知识管理模块获取意图相关知识。
可以理解的是，目前ENI架构上对于意图转译功能的实现存在两个可选方案：1)基于策略管理模块进行功能增强和2)作为架构的独立功能模块。
意图实施过程包括意图转译、执行以及维持。示例性地，在意图实施过程中，由意图转译模块负责意图转译，并且将转译的结果发送给策略管理模块，由策略管理模块进行意图执行以及维持。在意图执行及维持过程中，策略管理模块根据意图的达成情况调整意图的可执行命令。
此外，为了更好地实现意图的调用，对意图的表达式进行了定义，意图的表达式如下：
<intentExpression>:＝<IntentDrivenAction><IntentDrivenObject>
<IntentDrivenAction>:＝<IntentDrivenActionName><IntentDrivenActionProperties>
<IntentDrivenObject>:＝<IntentDrivenObjectName><IntentDrivenObjectProperties>
由如上表达式可以看出，一条完整的意图由两部分组成，意图动作(IntentDrivenAction)以及意图目标(IntentDrivenObject)。
意图系统从意图实施开始周期性地判断意图是否达成，当意图不能够达成时，调整意图对应的可执行命令。意图目标达成的检查周期受限于辅助系统的上报能力，由意图系统根据意图业务决定，最小可达1min，当辅助系统的上报周期较小时，意图系统在意图达成前会周期性的判断意图达成情况。其中，意图目标达成的检查周期与辅助系统的上报周期相同。
其中，本申请实施例中涉及的参考时长是指调整意图的可执行命令的最小时间间隔，即可执行命令开始被执行到下一次进行可执行命令调整所需要的最短时长。此外，该参考时长还可以称为期望时长(exepected time)，或者其他名称，本申请实施例对此不作限定。此外，需要说明的是，网络状态采集周期(即上报周期)小于参考时长。
为了减少意图系统在意图目标达成前乒乓切换意图可执行命令，提高意图系统达成意图目标的效率，减少资源的消耗，本申请实施例提供一种意图维持方法，如图2所示，该方法包括：
步骤200：状态感知模块获取意图对应的第一时长。
以下对状态感知模块获取第一时长的可能方式进行说明，可以理解的是，以下方式仅为举例不作为本申请实施例的限定。
方式1：第一时长可以由用于创建意图或实施意图或修改意图的消息携带。例如，意图创建者可以向意图系统发生创建意图消息，该消息包括第一时长，进一步地，第一时长可以由意图系统中的其他模块转发至状态感知模块。此时，第一时长为初始参考时长。
方式2：第一时长还可以配置在意图知识库中。例如，意图系统中的其他模块可以从意图知识库中获取第一时长并转发至状态感知模块。
进一步地，第一时长可以是指同种类型的意图调整可执行命令的平均时长。或者，第一时长还可以是根据模拟意图实施过程中意图目标达成所需的时长确定的。此时的第一时长为初始参考时长。
示例性地，在首次模拟意图执行时，意图系统记录意图目标的达成所需的时长，记为参考时长＝To。在第二次模拟意图的执行时，若To内达成意图的目标，则调整参考时长，即减少参考时长。若To内未达成意图的目标，则调整参考时长，即延长参考时长。具体的，若To内达成意图的目标，则可以利用公式T＝To+w*t对参考时长进行调整，此时t为负数，w为正数，t为意图的目标达成所用时长与To的差值。若To内未达成意图的目标，则在To的基础上再运行t’(例如，t’可以取To，具体的取值根据经验值设置)，如果经过t’达成了意图的目标，则可以利用公式T＝To+w*t对参考时长进行调整，t为超过To的时长，如果在t’时间内意图的目标未达成，则调整意图的可执行命令，并且利用公式T＝To+w*t对参考时长进行调整，此时t＝t’。
基于上述思路，在一些实施例中，可以在参考时长的变化趋势平稳后，确定一个参考时长作为第一时长。例如，假设预设浮动范围为40s，经过若干次模拟意图实施过程，参考时长的变化范围为4分45秒至5分15秒之间，此时参考时长的浮动范围为30s，落入预设浮动范围，则可以将第一时长设为5分钟。
在一些实施例中，在模拟意图实施过程达到预设次数时，将最后一次获得的参考时长作为第一时长。例如，假设预设次数为5次，当模拟意图实施过程达到5次时，将第5次 模拟意图实施过程后确定的参考时长作为第一时长。
步骤210：策略管理模块向状态感知模块发送第一信息，第一信息指示意图对应的第一执行命令开始执行。
在接收到第一信息后，状态感知模块监控第一可执行命令的执行时长。若第一可执行命令的执行时长到达第一时长且意图的目标未达成，则执行步骤220A。若意图的目标在第一时长到达前达成，则执行步骤220B。示例性地，在接收到第一信息后状态感知模块开始针对第一可执行命令的执行时长进行计时。
步骤220A：在第一可执行命令的执行时长到达第一时长且意图的目标未达成时，状态感知模块确定第二时长，第二时长大于第一时长。
在一示例中，状态感知模块可以自身确定第二时长。
在一示例中，状态感知模块可以向知识管理模块发送第四信息，第四信息指示第一可执行命令的执行时长到达第一时长且意图的目标未达成。知识管理模块在接收到第四信息后确定第二时长，并向状态感知模块发送第五信息，第五信息包括第二时长。
在一示例中，状态感知模块不需要向知识管理模块发送第四信息，但在意图的目标在第一时长到达前达成时，状态感知模块向策略管理模块发送第三信息，第三信息指示第一可执行命令的执行时长，即实际意图目标达成所需的时长。知识管理模块在确定未收到第三信息时确定第二时长，则向状态管理模块发送第五信息，第五信息包括第二时长。
此外，知识管理模块还可以根据第二时长更新同种类型的意图调整可执行命令的平均时长。
可以理解的是，上述仅以知识管理模块为例进行说明，还可以是意图系统中的其他模块确定第二时长并将第二时长通知给状态感知模块，本申请实施例对此不作限定。
进一步地，状态感知模块或知识管理模块可以采用如下公式确定第二时长：
Time_update＝Time_initial+△，△可以为预设值，△大于0。
Time_update表示第二时长，Time_initial表示第一时长，例如，△＝60s。
需要说明的是，状态感知模块或知识管理模块可以采用其他公式确定第二时长，本申请实施例对此不作限定。
步骤230A：状态感知模块向策略管理模块发送第二信息，第二信息指示第二时长。
步骤240A：在第一可执行命令的执行时长到达第一时长且意图的目标未达成时，策略管理模块将第一可执行命令调整为第二可执行命令。
可以理解的是，本申请实施例不限定步骤230A和步骤240A的执行顺序。
其中，策略管理模块需要首先获取意图对应的第一时长，策略管理模块获取第一时长的具体方式可以参考步骤200，重复之处不再赘述。
策略管理模块根据意图的目标是否在第一时长内达成决定是否需要调整第一可执行命令。若第一可执行命令的执行时长到达第一时长且意图的目标未达成，即意图的目标在第一时长内未达成，则策略管理模块需要调整可执行命令，即将意图的第一可执行命令调整为意图对应的第二可执行命令。需要说明的是，一个意图可以对应多个可执行命令，这里的多个可执行命令第一可执行命令和第二可执行命令。若意图的目标在第一时长到达前达成，则策略管理模块不需要调整可执行命令。
此外，策略管理模块向状态管理模块发送用于指示第二执行命令开始执行的消息。结合步骤230A，策略管理模块根据意图的目标是否在第二时长内达成决定是否需要调整第二 可执行命令，进入新的循环过程。
可以理解的是，以下仅以切换可执行命令为例进行说明，本申请实施例还适用于切换规则或者切换子意图。例如，在第一规则的执行时长到达第一时长且意图的目标未达成时，策略管理模块将第一规则调整为第二规则。或者，在第一子意图的执行时长到达第一时长且第一子意图的目标未达成时，策略管理模块将第一子意图调整为第二子意图。
步骤220B：在意图的目标在第一时长到达前达成时，第一可执行命令的执行时长小于第一时长，状态感知模块确定第三时长，第三时长小于第一时长。
其中，在一示例中，状态感知模块可以自身确定第三时长。
在另一示例中，状态感知模块可以向知识管理模块发送第六信息，第六信息指示第一可执行命令的执行时长。状态感知模块接收来自于知识管理模块的第七信息，第七信息指示第三时长。此外，知识管理模块还可以根据第三时长更新同种类型的意图调整可执行命令的平均时长。
可以理解的是，上述仅以知识管理模块为例进行说明，还可以是意图系统中的其他模块确定第三时长并将第三时长通知给状态感知模块，本申请实施例对此不作限定。
示例性地，第三时长为第一可执行命令的执行时长。
示例性地，状态感知模块或知识管理模块可以采用如下公式确定第三时长：
Time_update＝Time_initial+w*t，其中Time_initial为第二时长，w为权重值，例如，w＝1，t为第一可执行命令的执行时长与第一时长的差值。
需要说明的是，状态感知模块或知识管理模块可以采用其他公式确定第三时长，本申请实施例对此不作限定。
步骤230B：状态感知模块向策略管理模块发送第三信息，第三信息指示第三时长。
进一步地，策略管理模块根据意图的目标是否在第三时长内达成决定是否需要调整第一可执行命令。
需要说明的是，步骤220B和步骤230B为可选的步骤。若意图的目标在第一时长到达前达成，则状态感知模块可以不需要确定第三时长，策略管理模块继续根据第一时长判断是否调整第一可执行命令。
采用上述方法可以实现减少意图系统在意图目标达成前乒乓切换意图可执行命令，提高意图系统达成意图目标的效率，减少资源的消耗。
以下结合示例对本申请实施例进行详细说明。
实施例1：如图3所示为意图转译及执行的流程示意图之一。初始参考时长由意图创建者下发。
步骤1：意图创建者向数据读取和规范化模块发送新策略(newPolicy)消息，该新策略消息包括初始参考时长。其中，初始参考时长可以是指图2所示实施例中的第一时长。初始参考时长又可称为参考时长的初始值。
示例性地，意图创建者可以为用户或应用或运营支持系统(operations support system,OSS)或商业支持系统(business support system，BSS)。
步骤2：数据读取和规范化模块对新策略消息以及相关数据进行归一化，并且将归一化后的数据发送给知识管理模块。其中，归一化后的数据也包括初始参考时长。
步骤3：知识管理模块根据归一化后的数据更新意图知识库。
步骤4：知识管理模块向意图转译模块发送转译策略(translatePolicy)请求消息。转译请求消息包括初始参考时长。
步骤5：意图转译模块为意图策略(Intent Policy)分配标识。
步骤6：意图转译模块向知识管理模块发送标识更新请求，标识更新请求用于更新意图知识库中的意图标识。
步骤7：意图转译模块向输出生成和非规范化模块发送响应消息，响应消息用于告知意图已经开始被处理。
步骤8：输出生成和非规范化模块向意图创建者发送响应消息，响应消息用于告知意图已经开始被处理。
此外，意图转译模块还需进行意图语法词法检查。如果语法词法检查有误，意图转译模块向意图创建者发送错误反馈信息。意图转译模块向知识管理模块发送意图错误信息。如果意图语法词法检查无误，则继续执行步骤9。
步骤9：意图转译模块提取意图关键词。
步骤10：意图转译模块向知识管理模块发送意图关键词更新消息。
步骤11：意图转译模块向知识管理模块发送意图知识获取请求。
步骤12：知识管理模块向意图转译模块发送意图知识。
步骤13：意图转译模块将意图知识进行编排确定意图转译结果。
其中，示例性地，意图知识可能包括意图目标信息，网络状态参数的名称和与意图策略相关的网络元素的标识符等。
步骤14：意图转译模块向知识管理模块发送更新意图策略实例上下文(updPolsemantics)消息。
步骤15：意图转译模块向上下文感知模块发送意图策略转译完成(polComplete)消息。意图策略转译完成(polComplete)消息包括初始参考时长和意图转译结果。
步骤16：上下文感知模块对网络状态进行判断，并且向状态感知模块发送网络状态告知完成(ctxComplete)消息。网络状态告知完成(ctxComplete)消息包括初始参考时长和意图转译结果。
步骤17：状态感知模块对意图的目标是否达成进行判断，并向模型驱动工程模块发送网络状态告知完成消息。网络状态告知完成(ctxComplete)消息包括初始参考时长和意图转译结果。
对应上述步骤200，状态感知模块获取初始参考时长，为步骤20做准备。
步骤18：模型驱动工程模块根据网络状态告知完成消息将意图转译结果转化为可以被意图作用实体(即辅助系统)理解的形式，并且将其更新到知识管理模块中，并且向策略管理模块发送模式驱动工程完成(mdeComplete)消息。模式驱动工程完成(mdeComplete)消息包括初始参考时长和意图转译结果。
步骤19：策略管理模块对意图转译结果进行转换以及决策，得到的可执行命令发送给输出生成和非规范化模块。
这里的可执行命令可以对应上述图2所示实施中的第一可执行命令。
步骤20：策略管理模块向状态感知模块发送意图执行开始(exeBegin)消息，意图执行开始消息指示意图的可执行命令开始执行。
步骤21：状态感知模块根据接收到的意图执行开始消息开始计时。
如果在参考时长内达成了意图的目标，状态感知模块记录可执行命令的执行时长。如果在参考时长内未达成意图的目标，状态感知模块可以记录可执行命令的执行时长到达参考时长且意图的目标未达成，或者不作记录。
步骤22：输出生成和非规范化模块将可执行命令发送给被意图作用实体(即辅助系统)。
采用上述实施例可以实现在意图首次下发时携带参考时长，能够减少意图在达成目标之前调整意图可执行命令的概率，减少资源的消耗。
实施例2：如图4所示为意图转译及执行的流程示意图之二。初始参考时长存储在意图知识库中。
步骤1：意图创建者向数据读取和规范化模块发送新策略(newPolicy)消息，该新策略消息包括参考时长。
示例性地，意图创建者可以为用户或应用或OSS或BSS。
步骤2：数据读取和规范化模块对新策略消息以及相关数据进行归一化，并且将归一化后的数据发送给知识管理模块。其中，归一化后的数据也包括参考时长。
步骤3：知识管理模块根据归一化后的数据更新意图知识库。
步骤4：知识管理模块向意图转译模块发送转译策略(translatePolicy)请求消息。转译请求消息包括参考时长。
步骤5：意图转译模块为意图策略(Intent Policy)分配标识。
步骤6：意图转译模块向知识管理模块发送标识更新请求，标识更新请求用于更新意图知识库中的意图标识。
步骤7：意图转译模块向输出生成和非规范化模块发送响应消息，响应消息用于告知意图已经开始被处理。
步骤8：输出生成和非规范化模块向意图创建者发送响应消息，响应消息用于告知意图已经开始被处理。
此外，意图转译模块还需进行意图语法词法检查。如果语法词法检查有误，意图转译模块向意图创建者发送错误反馈信息。意图转译模块向知识管理模块发送意图错误信息。此外，如果意图语法词法检查无误，则继续执行步骤9。
步骤9：意图转译模块提取意图关键词。
步骤10：意图转译模块向知识管理模块发送意图关键词更新消息。
步骤11：意图转译模块向知识管理模块发送意图知识获取请求。
步骤12：知识管理模块向意图转译模块发送返回意图(returnIntent)消息，该消息携带初始参考时长和意图知识。
初始参考时长保存在意图知识库中。此时，初始参考时长可以为意图的类型对应的参考时长，即一类意图的参考时长。
步骤13：意图转译模块将意图知识进行编排确定意图转译结果。意图知识可能包括意图目标信息，网络状态参数的名称和与意图策略相关的网络元素的标识符等。
步骤14：意图转译模块向知识管理模块发送更新意图策略实例上下文(updPolsemantics)消息。
步骤15：意图转译模块向上下文感知模块发送意图策略转译完成(polComplete)消息。意图策略转译完成(polComplete)消息包括初始参考时长和意图转译结果。
步骤16：上下文感知模块对网络状态进行判断，并且向状态感知模块发送网络状态告 知完成(ctxComplete)消息。网络状态告知完成(ctxComplete)消息包括初始参考时长和意图转译结果。
步骤17：状态感知模块对意图的目标是否达成进行判断，并向模型驱动工程模块发送网络状态告知完成消息。网络状态告知完成(ctxComplete)消息包括初始参考时长和意图转译结果。
步骤18：模型驱动工程模块根据网络状态告知完成消息将意图转译结果转化为可以被意图作用实体(即辅助系统)理解的形式，并且将其更新到知识管理模块中，并且向策略管理模块发送模式驱动工程完成(mdeComplete)消息。模式驱动工程完成(mdeComplete)消息包括初始参考时长和意图转译结果。
步骤19：策略管理模块对意图转译结果进行转换以及决策，得到的可执行命令或者规则发送给输出生成和非规范化模块。
步骤20：策略管理模块向状态感知模块发送意图执行开始(exeBegin)消息，意图执行开始消息指示意图的可执行命令开始执行。
步骤21：状态感知模块根据接收到的意图执行开始消息开始计时。
如果在参考时长内达成了意图的目标，状态感知模块记录可执行命令的执行时长。如果在参考时长内未达成意图的目标，状态感知模块记录可执行命令的执行时长到达参考时长且意图的目标未达成。
步骤22：输出生成和非规范化模块将可执行命令或者规则发送给被意图作用实体(即辅助系统)。
采用上述实施例，在参考时长保存在意图知识库，能够减少意图在达成目标之前调整意图可执行命令的概率，减少资源的消耗。
实施例3：如图5所示为意图维持流程示意图之一。本实施例中知识管理模块可以确定更新后的参考时长。
步骤1：上下文感知模块向输出生成和非规范化模块发送性能数据订阅请求。
步骤2：输出生成和非规范化模块向辅助系统发送性能数据订阅请求。
步骤3：辅助系统向数据读取和规范化模块发送性能数据。
步骤4：数据读取和规范化模块对性能数据进行归一化处理，并将处理后的数据发送给知识管理模块。
步骤5：知识管理模块向上下文感知模块发送策略和性能数据(polAndPerformanceData)，策略和性能数据包括更新后的参考时长。
其中，在未达到参考时长且意图的目标达成时，状态感知模块可以向知识管理模块发送更新策略情景(updPolSA)消息,该消息包括测量时长，测量时长为实际测量得到的意图目标达成时间，即在参考时长内达成了意图的目标时，可执行命令的执行时长。知识管理模块根据测量时长确定更新后的参考时长。可以理解的是，这里的测量时长可以对应图2所示实施例中的第一可执行命令的执行时长。此时，更新后的参考时长可以对应图2所示实施例中的第三时长。
在一示例中，更新后的参考时长的计算公式为：
Time_update＝Time_initial+w*t，其中，Time_update为更新后的参考时长，Time_initial为状态感知模块中记录的参考时长，即最近一次更新的参考时长或初始参考时长。w为权重值，例如，w＝1，t为测量时长与Time_initial的差值。
在另一示例中，可以直接定义更新后的参考时长等于测量时长。
本申请实施例不限定状态感知模块向知识管理模块发送测量时长的时机。示例性地，状态感知模块在接收到来自于上下文感知模块的网络状态告知完成消息后，即步骤6后，状态感知模块向知识管理模块发送更新策略情景消息，该消息包括测量时长。此时的测量时长为前一轮可执行命令执行的执行时长。
在达到参考时长且意图的目标未达成时，状态感知模块可以向知识管理模块发送updPolSA,该消息指示达到参考时长且意图的目标未达成，知识管理模块根据该消息确定更新后的参考时长。示例性地，Time_update＝Time_initial+△，Time_update为更新后的参考时长，Time_initial为初始参考时长，△可以为预设值，△大于0。此时，更新后的参考时长可以对应图2所示实施例中的第二时长。
此外，在达到参考时长且意图的目标未达成时，状态感知模块可以不向知识管理模块发送消息。知识管理模块若未收到状态感知模块发送的消息则确定达到参考时长且意图的目标未达成，知识管理模块确定更新后的参考时长。
需要说明的是，当意图知识库保存意图的初始参考时长时，即对应于实施例2的场景，知识管理模块还可以根据更新后的参考时长更新意图系统中属于同一类的意图的初始参考时长。示例性地，T＝(w1*T1+w2*T2+w3*T3+……+wn*Tn)/n，其中n为意图系统中属于同一类的意图(例如提升带宽类意图)且处于激活状态的意图的个数，T1，T2，……Tn等为n个意图分别对应的初始参考时长。w1、w2……wn为T1，T2，……Tn分别对应的权重。
步骤6：上下文感知模块向状态感知模块发送网络状态告知完成消息，网络状态告知完成消息包括更新后的参考时长。
步骤7：状态感知模块向模型驱动工程模块发送网络状态告知完成消息，网络状态告知完成消息包括更新后的参考时长。
步骤8：模型驱动工程模块向策略管理模块发送模式驱动工程完成消息,模式驱动工程完成消息包括更新后的参考时长。
步骤9：策略管理模块根据更新后的参考时长和意图目标的达成情况判断是否要调整意图的可执行命令。若未达到更新后的参考时长且意图的目标达成，则不需要调整意图的可执行命，流程结束。若达到更新后的参考时长且意图的目标未达成，则需要调整意图的可执行命令，执行步骤10。
步骤10：策略管理模块将调整后的可执行命令发送给输出生成和非规范化模块。
步骤11：输出生成和非规范化模块将调整后的可执行命令发送给辅助系统。
步骤12：策略管理模块向状态感知模块发送意图执行开始消息，通知状态感知模块开始新一轮的可执行命令的执行时间计时。
采用上述实施例可以实现由知识管理模块对参考时长进行更新。
实施例4：如图6所示为意图维持流程示意图之二。本实施例中状态感知模块可以确定更新后的参考时长。
步骤1：上下文感知模块向输出生成和非规范化模块发送性能数据订阅请求。
步骤2：输出生成和非规范化模块向辅助系统发送性能数据订阅请求。
步骤3：辅助系统向数据读取和规范化模块发送性能数据。
步骤4：数据读取和规范化模块对性能数据进行归一化处理，并将处理后的数据发送给知识管理模块。
步骤5：知识管理模块对性能数据进行存储，并且将性能数据以及意图上下文信息送给上下文感知模块，其中，意图上下文信息是在意图转译以及执行阶段生成的。
步骤6：上下文感知模块根据性能数据判断网络状态，将网络状态信息增加至上下文信息中，并且将意图上下文信息更新到知识管理模块中，且将更新后的上下文信息发送给状态感知模块。
步骤7：状态感知模块确定更新后的参考时长。
在未达到参考时长且意图的目标达成时，状态感知模块根据实际测量得到的意图目标达成时间，即在参考时长内达成了意图的目标时，可执行命令的执行时长(以下简称测量时长)，计算更新后的参考时长。在一示例中，更新后的参考时长的计算公式为：
Time_update＝Time_initial+w*t，其中，Time_update为更新后的参考时长，Time_initial为状态感知模块中记录的参考时长，即最近一次更新的参考时长或初始参考时长，w为权重值，例如，w＝1，t为测量时长与Time_initial的差值。
在另一示例中，可以定义更新后的参考时长等于测量时长。
在达到参考时长且意图的目标未达成时，状态感知模块确定更新后的参考时长。示例性地，Time_update＝Time_initial+△，其中，Time_update为更新后的参考时长，Time_initial为状态感知模块中记录的参考时长，即最近一次更新的参考时长或初始参考时长，△可以为预设值，△大于0。
步骤8：状态感知模块向模型驱动工程模块发送网络状态告知完成消息，网络状态告知完成消息包括更新后的参考时长。
步骤9：模型驱动工程模块向策略管理模块发送模式驱动工程完成消息,驱动工程完成消息包括更新后的参考时长。
步骤10：策略管理模块根据更新后的参考时长和意图目标的达成情况判断是否要调整意图的可执行命令。若未达到更新后的参考时长且意图的目标达成，则不需要调整意图的可执行命，流程结束。若达到更新后的参考时长且意图的目标未达成，则需要调整意图的可执行命令，执行步骤11。
步骤11：策略管理模块将调整后的可执行命令发送给输出生成和非规范化模块。
步骤12：输出生成和非规范化模块将调整后的可执行命令发送给辅助系统。
步骤13：策略管理模块向状态感知模块发送意图执行开始消息，通知状态感知模开始新一轮的可执行命令的执行时间计时。
采用上述实施例可以实现由知识管理模块对参考时长进行更新。
可以理解的是，为了实现上述实施例中功能，网络设备和终端设备包括了执行各个功能相应的硬件结构和/或软件模块。本领域技术人员应该很容易意识到，结合本申请中所公开的实施例描述的各示例的单元及方法步骤，本申请能够以硬件或硬件和计算机软件相结合的形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行，取决于技术方案的特定应用场景和设计约束条件。
图7和图8为本申请的实施例提供的可能的通信装置的结构示意图。这些通信装置可以用于实现上述方法实施例中状态感知模块或策略管理模块或知识管理模块的功能，因此 也能实现上述方法实施例所具备的有益效果。在本申请的实施例中，该通信装置可以是如图1所示的状态感知模块或策略管理模块或知识管理模块，还可以是应用于状态感知模块或策略管理模块或知识管理模块的芯片。
如图7所示，通信装置700包括处理单元710和收发单元720。当通信装置700用于实现上述图2、图3、图4、图5或图6中所示的方法实施例中状态感知模块的功能时，收发单元720用于获取意图对应的第一时长，接收来自于策略管理模块的第一信息，第一信息指示意图对应的第一执行命令开始执行。
处理单元710用于在第一可执行命令的执行时长到达第一时长且意图的目标未达成时，状态感知模块确定第二时长，第二时长大于第一时长。
收发单元720用于向策略管理模块发送第二信息，第二信息指示第二时长。
当通信装置700用于实现图2、图3、图4、图5或图6所示的方法实施例中策略管理模块的功能时：收发单元720用于获取意图对应的第一时长，向状态感知模块发送第一信息，第一信息指示意图对应的第一可执行命令开始执行；
处理单元710用于在第一可执行命令的执行时长到达第一时长且意图的目标未达成时，将第一可执行命令调整为意图对应的第二可执行命令；
收发单元720用于接收来自于状态感知模块的第二信息，第二信息指示第二时长，第二时长大于第一时长。
当通信装置700用于实现图2、图3、图4、图5或图6所示的方法实施例中知识管理模块的功能时：收发单元720用于接收来自于状态感知模块的第四信息，第四信息指示意图的第一可执行命令的执行时长到达第一时长且意图的目标未达成；
处理单元710用于确定第二时长，第二时长大于第一时长；
收发单元720用于向状态感知模块发送第五信息，第五信息指示第二时长。
当通信装置700用于实现图2、图3、图4、图5或图6所示的方法实施例中知识管理模块的功能时：收发单元720用于接收来自于状态感知模块的第六信息，第六信息指示意图的目标被达成时第一可执行命令的执行时长；
处理单元710用于根据第一可执行命令的执行时长确定第三时长，第三时长小于第一时长；
收发单元720用于向状态感知模块发送第七信息，第七信息指示第三时长。
有关上述处理单元710和收发单元720更详细的描述可以直接参考图2、图3、图4、图5或图6所示的方法实施例中相关描述直接得到，这里不加赘述。
如图8所示，通信装置800包括处理器810和接口电路820。处理器810和接口电路820之间相互耦合。可以理解的是，接口电路820可以为收发器或输入输出接口。可选的，通信装置800还可以包括存储器830，用于存储处理器810执行的指令或存储处理器810运行指令所需要的输入数据或存储处理器810运行指令后产生的数据。
当通信装置800用于实现图2、图3、图4、图5或图6所示的方法时，处理器810用于实现上述处理单元710的功能，接口电路820用于实现上述收发单元720的功能。
可以理解的是，本申请的实施例中的处理器可以是中央处理单元(Central Processing Unit，CPU)，还可以是其它通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现场可编程门阵列(Field Programmable Gate Array，FPGA)或者其它可编程逻辑器件、晶体管逻辑器件，硬件部件 或者其任意组合。通用处理器可以是微处理器，也可以是任何常规的处理器。
本申请的实施例中的方法步骤可以通过硬件的方式来实现，也可以由处理器执行软件指令的方式来实现。软件指令可以由相应的软件模块组成，软件模块可以被存放于随机存取存储器(Random Access Memory，RAM)、闪存、只读存储器(Read-Only Memory，ROM)、可编程只读存储器(Programmable ROM，PROM)、可擦除可编程只读存储器(Erasable PROM，EPROM)、电可擦除可编程只读存储器(Electrically EPROM，EEPROM)、寄存器、硬盘、移动硬盘、CD-ROM或者本领域熟知的任何其它形式的存储介质中。一种示例性的存储介质耦合至处理器，从而使处理器能够从该存储介质读取信息，且可向该存储介质写入信息。当然，存储介质也可以是处理器的组成部分。处理器和存储介质可以位于ASIC中。另外，该ASIC可以位于网络设备或终端设备中。当然，处理器和存储介质也可以作为分立组件存在于网络设备或终端设备中。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机程序或指令。在计算机上加载和执行所述计算机程序或指令时，全部或部分地执行本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、网络设备、用户设备或者其它可编程装置。所述计算机程序或指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机程序或指令可以从一个网站站点、计算机、服务器或数据中心通过有线或无线方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是集成一个或多个可用介质的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质，例如，软盘、硬盘、磁带；也可以是光介质，例如，数字视频光盘(digital video disc，DVD)；还可以是半导体介质，例如，固态硬盘(solid state drive，SSD)。
在本申请的各个实施例中，如果没有特殊说明以及逻辑冲突，不同的实施例之间的术语和/或描述具有一致性、且可以相互引用，不同的实施例中的技术特征根据其内在的逻辑关系可以组合形成新的实施例。
本申请中，“至少一个”是指一个或者多个，“多个”是指两个或两个以上。“和/或”，描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B的情况，其中A,B可以是单数或者复数。在本申请的文字描述中，字符“/”，一般表示前后关联对象是一种“或”的关系；在本申请的公式中，字符“/”，表示前后关联对象是一种“相除”的关系。
可以理解的是，在本申请的实施例中涉及的各种数字编号仅为描述方便进行的区分，并不用来限制本申请的实施例的范围。上述各过程的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定。

Claims (30)

  1. 一种意图维持方法，其特征在于，该方法包括：
    状态感知模块获取意图对应的第一时长；
    所述状态感知模块接收来自于策略管理模块的第一信息，所述第一信息指示所述意图对应的第一执行命令开始执行；
    在所述第一可执行命令的执行时长到达所述第一时长且所述意图的目标未达成时，所述状态感知模块确定第二时长，所述第二时长大于所述第一时长；
    所述状态感知模块向所述策略管理模块发送第二信息，所述第二信息指示所述第二时长。
  2. 如权利要求1所述的方法，其特征在于，还包括：
    在所述意图的目标在所述第一时长到达前达成时，所述第一可执行命令的执行时长小于所述第一时长，所述状态感知模块确定第三时长，所述第三时长小于所述第一时长；
    所述状态感知模块向所述策略管理模块发送第三信息，所述第三信息指示所述第三时长。
  3. 如权利要求2所述的方法，其特征在于，所述第三时长是根据所述第一可执行命令的执行时长确定的。
  4. 如权利要求1所述的方法，其特征在于，所述状态感知模块确定第二时长，包括：
    所述状态感知模块向知识管理模块发送第四信息，所述第四信息指示所述第一可执行命令的执行时长到达所述第一时长且所述意图的目标未达成；
    所述状态感知模块接收来自于所述知识管理模块的第五信息，所述第五信息指示所述第二时长。
  5. 如权利要求2或3所述的方法，其特征在于，所述状态感知模块确定第三时长，包括：
    所述状态感知模块向知识管理模块发送第六信息，所述第六信息指示所述第一可执行命令的执行时长；
    所述状态感知模块接收来自于所述知识管理模块的第七信息，所述第七信息指示所述第三时长。
  6. 如权利要求1-5任一项所述的方法，其特征在于，所述第一时长由第八信息携带，所述第八信息用于创建或实施或修改所述意图；
    或者，所述第一时长配置在意图知识库中。
  7. 如权利要求1-6任一项所述的方法，其特征在于，所述第一时长是根据模拟所述意图实施过程中意图目标达成所需的时长确定的。
  8. 一种意图维持方法，其特征在于，该方法包括：
    策略管理模块获取意图对应的第一时长；
    所述策略管理模块向状态感知模块发送第一信息，所述第一信息指示意图对应的第一可执行命令开始执行；
    在所述第一可执行命令的执行时长到达所述第一时长且所述意图的目标未达成时，所述策略管理模块将所述第一可执行命令调整为所述意图对应的第二可执行命令；
    所述策略管理模块接收来自于所述状态感知模块的第二信息，所述第二信息指示所述 第二时长，所述第二时长大于所述第一时长。
  9. 如权利要求8所述的方法，其特征在于，还包括：
    在所述意图的目标在所述第一时长到达前达成时，所述第一可执行命令的执行时长小于所述第一时长，所述策略管理模块接收来自于所述状态感知模块的第三信息，所述第三信息指示所述第三时长，所述第三时长小于所述第一时长。
  10. 如权利要求9所述的方法，其特征在于，所述第三时长是根据所述第一可执行命令的执行时长确定的。
  11. 如权利要求8-10任一项所述的方法，其特征在于，所述第一时长由第八信息携带，所述第八信息用于创建或实施或修改所述意图；
    或者，所述第一时长配置在意图知识库中。
  12. 如权利要求8-11任一项所述的方法，其特征在于，所述第一时长是根据模拟所述意图实施过程中意图目标达成所需的时长确定的。
  13. 一种意图维持方法，其特征在于，该方法包括：
    知识管理模块接收来自于状态感知模块的第四信息，第四信息指示意图的第一可执行命令的执行时长到达第一时长且所述意图的目标未达成；
    所述知识管理模块确定第二时长，所述第二时长大于所述第一时长；
    所述知识管理模块向所述状态感知模块发送第五信息，所述第五信息指示所述第二时长。
  14. 如权利要求13所述的方法，其特征在于，还包括：
    所述知识管理模块根据所述第二时长更新与所述意图的类型相同的意图所对应的第一时长平均值。
  15. 一种意图维持方法，其特征在于，该方法包括：
    知识管理模块接收来自于状态感知模块的第六信息，第六信息指示意图的目标被达成时第一可执行命令的执行时长；
    所述知识管理模块根据所述第一可执行命令的执行时长确定第三时长，所述第三时长小于所述第一时长；
    所述知识管理模块向所述状态感知模块发送第七信息，所述第七信息指示所述第三时长。
  16. 如权利要求15所述的方法，其特征在于，还包括：
    所述知识管理模块根据所述第三时长更新与所述意图的类型相同的意图所对应的第一时长平均值。
  17. 一种意图维持装置，其特征在于，该装置包括：
    收发单元，用于获取意图对应的第一时长；接收来自于策略管理模块的第一信息，所述第一信息指示所述意图对应的第一执行命令开始执行；
    处理单元，用于在所述第一可执行命令的执行时长到达所述第一时长且所述意图的目标未达成时，确定第二时长，所述第二时长大于所述第一时长；
    所述收发单元，用于向所述策略管理模块发送第二信息，所述第二信息指示所述第二时长。
  18. 如权利要求17所述的装置，其特征在于，所述处理单元，用于在所述意图的目标在所述第一时长到达前达成时，所述第一可执行命令的执行时长小于所述第一时长，确定 第三时长，所述第三时长小于所述第一时长；
    所述收发单元，用于向所述策略管理模块发送第三信息，所述第三信息指示所述第三时长。
  19. 如权利要求18所述的装置，其特征在于，所述第三时长是根据所述第一可执行命令的执行时长确定的。
  20. 如权利要求17所述的装置，其特征在于，所述收发单元，用于向知识管理模块发送第四信息，所述第四信息指示所述第一可执行命令的执行时长到达所述第一时长且所述意图的目标未达成；
    所述收发单元，用于接收来自于所述知识管理模块的第五信息，所述第五信息指示所述第二时长。
  21. 如权利要求18或19所述的装置，其特征在于，所述收发单元，用于向知识管理模块发送第六信息，所述第六信息指示所述第一可执行命令的执行时长；
    所述收发单元，用于接收来自于所述知识管理模块的第七信息，所述第七信息指示所述第三时长。
  22. 一种意图维持装置，其特征在于，该装置包括：
    收发单元，用于获取意图对应的第一时长；
    所述收发单元，用于向状态感知模块发送第一信息，所述第一信息指示意图对应的第一可执行命令开始执行；
    处理单元，用于在所述第一可执行命令的执行时长到达所述第一时长且所述意图的目标未达成时，将所述第一可执行命令调整为所述意图对应的第二可执行命令；
    所述收发单元，用于接收来自于所述状态感知模块的第二信息，所述第二信息指示所述第二时长，所述第二时长大于所述第一时长。
  23. 如权利要求22所述的装置，其特征在于，所述收发单元，用于在所述意图的目标在所述第一时长到达前达成时，所述第一可执行命令的执行时长小于所述第一时长，接收来自于所述状态感知模块的第三信息，所述第三信息指示所述第三时长，所述第三时长小于所述第一时长。
  24. 如权利要求23所述的装置，其特征在于，所述第三时长是根据所述第一可执行命令的执行时长确定的。
  25. 一种意图维持装置，其特征在于，该装置包括：
    收发单元，用于接收来自于状态感知模块的第四信息，第四信息指示意图的第一可执行命令的执行时长到达第一时长且所述意图的目标未达成；
    处理单元，用于确定第二时长，所述第二时长大于所述第一时长；
    所述收发单元，用于向所述状态感知模块发送第五信息，所述第五信息指示所述第二时长。
  26. 如权利要求25所述的装置，其特征在于，所述处理单元，用于根据所述第二时长更新与所述意图的类型相同的意图所对应的第一时长平均值。
  27. 一种意图维持装置，其特征在于，该装置包括：
    收发单元，用于接收来自于状态感知模块的第六信息，第六信息指示意图的目标被达成时第一可执行命令的执行时长；
    所述处理单元，用于根据所述第一可执行命令的执行时长确定第三时长，所述第三时 长小于所述第一时长；
    所述收发单元，用于向所述状态感知模块发送第七信息，所述第七信息指示所述第三时长。
  28. 一种通信装置，其特征在于，包括用于执行如权利要求1至16中的任一项所述方法的模块。
  29. 一种通信装置，其特征在于，包括处理器和接口电路，所述接口电路用于接收来自所述通信装置之外的其它通信装置的信号并传输至所述处理器或将来自所述处理器的信号发送给所述通信装置之外的其它通信装置，所述处理器通过逻辑电路或执行代码指令用于实现如权利要求1至16中任一项所述的方法。
  30. 一种计算机可读存储介质，其特征在于，所述存储介质中存储有计算机程序或指令，当所述计算机程序或指令被通信装置执行时，实现如权利要求1至16中任一项所述的方法。
PCT/CN2021/125201 2020-11-18 2021-10-21 一种意图维持方法及装置 WO2022105522A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP21893670.6A EP4227827A4 (en) 2020-11-18 2021-10-21 METHOD AND DEVICE FOR MAINTAINING INTENTION
US18/320,104 US20230291651A1 (en) 2020-11-18 2023-05-18 Intent maintenance method and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011295591.1A CN114519134A (zh) 2020-11-18 2020-11-18 一种意图维持方法及装置
CN202011295591.1 2020-11-18

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/320,104 Continuation US20230291651A1 (en) 2020-11-18 2023-05-18 Intent maintenance method and apparatus

Publications (1)

Publication Number Publication Date
WO2022105522A1 true WO2022105522A1 (zh) 2022-05-27

Family

ID=81594747

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/125201 WO2022105522A1 (zh) 2020-11-18 2021-10-21 一种意图维持方法及装置

Country Status (4)

Country Link
US (1) US20230291651A1 (zh)
EP (1) EP4227827A4 (zh)
CN (1) CN114519134A (zh)
WO (1) WO2022105522A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101280729A (zh) * 2007-04-03 2008-10-08 丰田自动车株式会社 节能运行促进系统及方法
US20150135085A1 (en) * 2013-11-08 2015-05-14 Kavaanu, Inc. System and method for activity management presentation
CN107133345A (zh) * 2017-05-22 2017-09-05 北京百度网讯科技有限公司 基于人工智能的交互方法和装置
CN111858890A (zh) * 2020-07-23 2020-10-30 平安证券股份有限公司 基于机器人的咨询方法、咨询机器人、计算机设备及介质

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8824437B2 (en) * 2011-03-02 2014-09-02 Ricoh Company, Ltd. Wireless communications device, electronic apparatus, and methods for determining and updating access point
US10075450B2 (en) * 2015-05-29 2018-09-11 Rockwell Automation Technologies, Inc. One time use password for temporary privilege escalation in a role-based access control (RBAC) system
EP3557439A1 (en) * 2018-04-16 2019-10-23 Tata Consultancy Services Limited Deep learning techniques based multi-purpose conversational agents for processing natural language queries
US11727068B2 (en) * 2021-11-29 2023-08-15 Salesforce, Inc. Customizable intent driven application triggering system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101280729A (zh) * 2007-04-03 2008-10-08 丰田自动车株式会社 节能运行促进系统及方法
US20150135085A1 (en) * 2013-11-08 2015-05-14 Kavaanu, Inc. System and method for activity management presentation
CN107133345A (zh) * 2017-05-22 2017-09-05 北京百度网讯科技有限公司 基于人工智能的交互方法和装置
CN111858890A (zh) * 2020-07-23 2020-10-30 平安证券股份有限公司 基于机器人的咨询方法、咨询机器人、计算机设备及介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4227827A4 *

Also Published As

Publication number Publication date
EP4227827A1 (en) 2023-08-16
US20230291651A1 (en) 2023-09-14
EP4227827A4 (en) 2024-04-10
CN114519134A (zh) 2022-05-20

Similar Documents

Publication Publication Date Title
WO2022011862A1 (zh) 一种o-ran与mec的通信方法及系统
US20230297882A1 (en) Ml ue capability and inability
WO2019196813A1 (zh) 一种订阅服务的方法及装置
US20220014942A1 (en) Ml model management in o-ran
WO2021164562A1 (zh) 终端行为数据的获取方法、发送方法、装置及网络设备
JP2008537858A (ja) ワイヤレス端末においてトラップを管理及び監視するためのシステム及び方法
EP3780496B1 (en) Feature engineering programming method and apparatus
US20230224752A1 (en) Communication method, apparatus, and system
WO2021244377A1 (zh) 一种模型更新方法及装置
TW202234908A (zh) 一種通信方法和通信裝置
WO2022105522A1 (zh) 一种意图维持方法及装置
WO2022057564A1 (zh) 通信方法及装置、系统
US11909600B2 (en) Intent negotiation method and apparatus
WO2022042544A1 (zh) 一种网络性能数据订阅方法及装置
CN117151669A (zh) 基于工作流引擎执行时间的提醒方法及装置、电子设备
US20230156547A1 (en) Switching between a defined generation of stand alone and non-stand alone wireless deployment based on determined service requirements, network capabilities, and user equipment capabilities
CN114095367A (zh) 一种状态反馈方式的配置方法、状态反馈方法及装置
WO2023169097A1 (zh) 5g消息终端的识别方法、服务平台和存储介质
WO2023185711A1 (zh) 一种用于机器学习模型训练的通信方法及装置
US20240211775A1 (en) Communication network
WO2024093876A1 (zh) 一种通信方法及装置
WO2023213246A1 (zh) 模型选择方法、装置及网络侧设备
WO2023125679A1 (zh) 信息获取方法、装置及通信设备
US12035156B2 (en) Communication method and apparatus for plurality of administrative domains
WO2022165723A1 (zh) 一种通信的方法和装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21893670

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021893670

Country of ref document: EP

Effective date: 20230509

NENP Non-entry into the national phase

Ref country code: DE