CN109245916B - Intention-driven cloud access network system and method - Google Patents
Intention-driven cloud access network system and method Download PDFInfo
- Publication number
- CN109245916B CN109245916B CN201810929724.2A CN201810929724A CN109245916B CN 109245916 B CN109245916 B CN 109245916B CN 201810929724 A CN201810929724 A CN 201810929724A CN 109245916 B CN109245916 B CN 109245916B
- Authority
- CN
- China
- Prior art keywords
- intention
- network
- module
- resource
- service
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
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
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
Abstract
The invention belongs to the technical field of wireless communication networks, and discloses an intention-driven cloud access network system and method.A service application layer obtains a mobile network core application and a service appeal facing vertical industry application, and translation is completed through an intention northbound interface; an intention enabling layer builds a global dynamic management and arrangement control system based on intents by relying on a cloud network service plane, and realizes unified planning of service intents and unified scheduling of resources; the infrastructure layer forms a resource pool by utilizing a resource virtualization technology, and provides ubiquitous connection based on three 5G application scenes. The invention realizes network collaboration, resource sharing, interface opening and management intellectualization through the intent-driven cloud ubiquitous extreme intelligence access network, and constructs a reconfigurable resource management architecture. The invention also discloses an intention-driven method for realizing the cloud access network, which is used for cooperatively scheduling network resources and providing end-to-end service guarantee.
Description
Technical Field
The invention belongs to the technical field of wireless communication networks, and particularly relates to an intention-driven cloud access network system and method.
Background
The future wireless communication network has the characteristics of high function modularization, resource sharing, interface opening, intelligent management, capability opening and the like, and supports business to customize resources as required through a strong converged network. The Software Defined Network (SDN) is a novel network architecture, and the design concept of the SDN is to separate a control plane and a data forwarding plane of the network, so that programmable control bottom hardware is realized through a software platform in a centralized controller, and flexible network resource allocation according to needs is realized. In the SDN network, the network device is only responsible for pure data forwarding, and may adopt general hardware; the original operating system responsible for control is abstracted into an independent network operating system which is responsible for adapting to different service characteristics, and the communication among the network operating system, the service characteristics and the hardware equipment can be realized through programming.
Virtualization (Virtualization) is a resource management technology, which abstracts and converts various physical resources of a computer, such as a server, a network, a memory, a storage and the like, to present the resources, so as to break the barrier that the physical structures cannot be cut, and enable a user to apply the resources in a better way than the original configuration. The network is built on a cloud platform based on a general infrastructure, network element functions are virtualized, a control plane is highly centralized, a data plane is decentralized, and rapid deployment of the network elements according to requirements is facilitated. Due to the large infrastructure and the difference of the communication network in terms of physical infrastructure, protocol and performance, if a system under the current SDN architecture needs to maintain a plurality of virtual networks simultaneously, functions of application programs are not affected mutually while resources are shared, the operation of the programs is coordinated, and the resource utilization efficiency is improved. In addition, when making a cloud resource decision according to the service requirements of the cloud users, the cloud operator mainly relies on the skills and experience of the cloud operator. However, given the large number of cloud service orders, the wide variety of requirements, and the complexity of cloud configuration options, such manual decisions tend to be costly and error prone. In order to effectively manage the end-to-end network, applications need to be managed from a higher level of abstraction, rather than at the details of the low-level devices, such an abstraction policy that provides guidance for the network is often seen as an intent. Whereas the radio access network is the last kilometer from the application or user, it is necessary to explore the intendedly driven radio access networks.
The prior art is a method and system for network traffic characterization, monitoring and control based on business intent driven policies. The method includes (1) determining network requirements for at least one application; (2) dynamically determining a link suitable for data transmission to meet the network requirements according to a policy based at least in part on current network conditions; (3) and routing one or more application network data flows associated with the at least one application over the link. The method has the following defects: the method utilizes the strategy character string to express the business intention in the application network driving model, but only considers the representation of the intention, and does not consider whether the represented intention is put down in a single mode or a plurality of modes, so that the consistency check problem of the intention is caused.
Prior art two is a technique for obtaining and processing intent-based feedback. Authored content may be generated by utilizing one or more layouts. The generated layout may then be used to present the content to the author or other user. The author may also provide feedback to request adjustments or re-generation of the generated layout. The feedback may include overlay feedback that overlays the intent interpretation used in the layout and intent feedback; the intent feedback changes intent data for the content data or adds more intent data for the content data. The technology has the following defects: this approach enhances user interaction with the computing device by modifying the layout of the content based on user feedback, but the feedback mechanism is based on the language, methodology and transformation actions of the specific computer structural features, and is not robust and extensible for the feedback function.
Three prior art methods for provisioning a network communication domain include: (1) defining a network communication domain having a plurality of network forwarding elements; (2) detecting, by an intent-driven controller, one or more capabilities associated with a plurality of network forwarding elements; (3) assigning requirements to each communication service of a plurality of computer applications running in the network and assigning a weight to data traffic associated with each application; (4) selecting the computer application program with the highest weight and temporarily determining a path passing through the network forwarding element to meet the communication service requirement; (5) one or more network forwarding elements are configured to allow application data traffic to be forwarded through the network according to the temporarily determined network path. The method has the following defects: the management control function of the intention-driven controller in the method is not directed to the 5G RAN, on one hand, the controller is not combined with a management arrangement system, on the other hand, the technology is applicable to a network environment, focuses on a fixed network and is not migrated to a wireless communication network, and efficient, elastic and flexible management capability cannot be provided in the face of diversified task requirements in the 5G era.
In summary, the problems of the prior art are as follows:
the traditional point-to-point transmission has obvious limitation, the traditional information theory only considers the point-to-point transmission process, and a transmission motivation, transmission content and a transmission result are not discussed; virtualization or software defined networking introduces network configuration complexity, communication application intent is disjointed from the network; the network supply and demand highlight the segmentation, the user demand and the network supply cannot be aligned accurately, the network lacks agile and quick response capability, and intelligent and simple network management cannot be realized; network element equipment highlights low intelligence, the inherent design of the equipment such as a terminal and the like reduces the participation degree of a user, and the user experience of the user can not be realized; the network evolution is obvious in stagnation, the network structure is closed, fusion, expansion and evolution are difficult, and node evolution is disjointed with network structure evolution. Therefore, an elastic network structure and node functions are needed, and a network virtualization pipeline and a control mechanism are constructed, so that the working mode and the network operation state are dynamically reconstructed along with the change of an application request.
The difficulty and significance of solving the technical problems are as follows:
the continuous increase of the scale and complexity of the wireless network, the diversified development of user requirements and services, the closed and inflexible rigid network architecture, the division and inexplosiveness of key network technologies cannot meet the requirements of the next generation of networks in the aspects of intelligence simplicity, agility, expansibility, psychology and convergence. In order to break through the bottleneck faced by the current wireless network, the Shannon theory is considered again from the network dimension, the network architecture is innovated again, and the network technology is exploited again. The method is driven by signaling and data, the aim of flexible, quick and accurate network management and node interaction is realized by aiming at the user intention by the research intention drive, and the intention-driven wireless network has important scientific significance as the evolution direction of the next generation network.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides an intention-driven cloud access network system and method.
The present invention is achieved in such a way that an intention-driven cloud access network system includes:
the service application layer programs the bottom layer equipment through a programming interface provided by the intention enabling layer, abstractly encapsulates the capability of the network element, and opens the control right of the network to users and other service partners; and providing a management interface;
the intention enabling layer is used for analyzing and checking consistency of an intention flow translated by the application request through the intention northbound interface and processing the intention into a network intention capable of being executed by the current network resource; obtaining corresponding operation of specific resources in the network through a mapping algorithm of the intents and the resources; the system based on the management and arrangement of the intentions is adopted to realize the unified planning of the global application intentions, the unified scheduling of resources and the closed-loop arrangement mode to realize the life cycle management of the network elements;
and the infrastructure layer is used for independently providing the non-real-time functions CU-C and CU-U of one or more access systems to form an access aggregation layer and carrying out unified cooperative management on the DUs of one or more access systems.
The intention-driven clouded access network system further comprises: an intent engine module;
the intent engine module includes:
an intent parsing module for understanding and learning intent streams translated via the northbound interface;
the system comprises an intention checking module, a logic consistency checking module and a logic consistency checking module, wherein the intention checking module is used for solving the problem of a large amount of logic consistency caused by possible intention conflicts when network delay occurs or a plurality of application programs send intents at the same time;
the resource mapping module realizes the automatic acquisition of resource combination decision through rule matching;
a resource combination template, which is used for storing the resource combination template and defining the basic resource composition of each basic service;
the parameter calculation module is used for determining the configuration parameters of the virtual network element according to the service requirements and the resource constraints;
the autonomous learning module is used for constructing a parameter decision scheme model by adopting supervised learning based on empirical data;
and the optimization verification module is used for pre-verifying the validity and the correctness of the configuration scheme and simultaneously enabling the network to obtain the self-optimization and self-healing capabilities in the long-term feedback verification.
Another object of the present invention is to provide an intention-driven cloud access network method using the intention-driven cloud access network system, the intention-driven cloud access network method including the steps of:
step one, an intention northbound interface embodies an application intention through a declarative expression rule and deconstructs the application intention into a specific expression form of an action, an object, a modification and a type; the translated intent stream is entered into an intent-based management and orchestration system (IB-MANO) of the cloud network service plane through the northbound intent interface;
step two, the intention flow enters an intention analysis module in an intention engine to identify a request in the application intention, and rule matching is carried out on an operation indicator in the intention flow; based on a commitment theory, an intention consistency checking module is used for solving the problem of a large amount of logic consistency caused by possible intention conflict when network delay occurs or a plurality of application programs send intents at the same time;
step three, aiming at the analyzed request of the intention flow, determining the resource combination by combining the resource combination rule stored in the resource combination template library; analyzing the priority and conflict of the resource combination scheme, and determining the type of the virtual network element;
fourthly, the parameter calculation module generates configuration parameters according to the working load and the environmental information and by combining the historical parameter decision information learned in the self-learning module; the determined resource composition and parameters are generated in YAML format and transmitted to the optimization verification module;
comparing the obtained configuration parameters with the real-time network state and historical data in an optimization verification module to verify the correctness of parameter configuration; the data file is delivered to a virtual network function manager to complete the arrangement and management; after the application intents are realized, the optimization verification module continuously verifies the reliability of the intention realization and realizes data updating in cooperation with the self-learning module;
step six, access network side feedback information acquisition: the centralized management node establishes a network service overall data model by adopting technologies such as big data, machine learning and the like, transmits key index data to the self-learning module and the optimization verification module through a feedback link in real time, realizes scene prejudgment at the management node and carries out pre-allocation scheduling of resources.
Further, the conflict and priority analysis specifically includes:
(1) determining the network configuration attribute associated with the executed strategy behavior according to the input strategy event and the priority relation matrix; if for each network configuration attribute, a plurality of policies related to the network configuration attribute are adopted at the same time, a conflict occurs;
(2) according to the priority information defined in the priority matrix, executing the network configuration attribute associated with the policy with the highest priority, and refusing to execute other policies;
(3) the priority relation matrix is constructed by taking a network element as a unit, and the matrix describes the priority of the strategy behavior adopted to operate the relevant configuration attributes under the condition that the judgment condition of the strategy is met.
Further, the self-learning module adopts a supervision learning mechanism, a linear regression model and a deep neural network to realize the self-learning mechanism. In addition, the self-learning module defines an upper limit of the processing time in advance to ensure the timely realization of the application intention.
Another object of the present invention is to provide a method for generating an intention engine implementation from an intention flow to a resource action flow by applying the intention-driven cloud access network method, wherein the method for generating the intention engine implementation from the intention flow to the resource action flow comprises:
step one, analyzing intention and checking consistency;
(1) the intention analysis module divides a group of data of the intention stream into parts of speech and loads labels, and realizes rule matching for each label attached symbol; outputting the analyzed atom request in a JSON file; an intent flow is denoted as < action, object, modifier, type. >, an action is the most critical operation describing an intent, an object identifies the executor or recipient of the action, a modifier is used to materialize or parameterize a certain constraint, and a type is an identification of a connection to another service process;
(2) based on the commitment theory, analyzing different intention targets when the intention arrives, generating an optimized solution to meet the requirement, and issuing the solution to a lower layer to execute a conflict-free intention after the conflict is solved at an upper layer; the form of the commitment theory is expressed as:
wherein n is1Is a commitment person, n2Is the recipient of the commitment, pi is the type of commitment, describes the commitment characteristics; the type of commitment is divided into basis, cooperation, use and condition;
in the network, the commitment device and the commitment receiver can be various nodes in the network, and the relationship between the node pairs is characterized through the commitment theory; the set of operations intended to satisfy all commitments involving nodes or to maximally satisfy all node commitments involved;
and step two, obtaining a resource combination scheme by strategy mapping.
(1) The analyzed request of the intention flow is combined with the resource combination rule stored in the resource combination template library to realize the determination of the resource combination;
(2) analyzing the priority and conflict of the resource combination scheme, determining the type of the virtual network element, and analyzing the service level, the service reliability and the service safety;
and step three, calculating resource configuration parameters.
(1) The parameter calculation module generates configuration parameters according to the working load and the environmental information and by combining the historical parameter decision information learned in the self-learning module; the self-learning module adopts a supervised learning mechanism, namely a linear regression model and a deep neural network to realize the self-learning mechanism;
(2) the determined resource composition and parameters are generated in YAML format and transmitted to the optimization verification module;
step four, continuously performing closed-loop optimization verification;
(1) comparing the obtained configuration parameters with the real-time network state and the historical data in an optimization verification module to verify the correctness of the parameter configuration; the data file is delivered to a virtual network function manager to complete the arrangement and management;
(2) after the intention is realized, the optimization verification module continuously verifies the reliability of the intention realization and realizes data updating in cooperation with the self-learning module.
Another object of the present invention is to provide a wireless communication network system applying the intention-driven clouded access network method.
In summary, the advantages and positive effects of the invention are: compared with the traditional mode, the established intention-driven cloud access network system improves the usability and agility of the network, deepens the intention into the modules, improves the intelligent degree of each module in the network architecture, implants intelligent genes into the network, enables the network to understand the application intention, and simplifies the management and operation and maintenance of the network. The invention can bring practical gains to different role personnel in the network: for network management personnel, engineers may focus more on network services rather than specific micro-configurations when configuring an intent-driven, clouded access network, bringing network delivery closer to business needs. For a terminal user, the network automatically adjusts and adapts to different services, and better user experience is provided; for network service providers or equipment providers, the network configuration and maintenance work is simpler and labor-saving.
The invention can reduce the manual wireless network parameter configuration by 80 percent, reduce the network operation and maintenance Opex by 80 percent, achieve the autonomous solving rate of the network problems by 99 percent, improve the detection rate of unknown threats to 99 percent, improve the resource utilization efficiency on the whole, refine the granularity of resource control and realize the wireless communication network system which is flexible, reconfigurable, evolvable, open, programmable and elastically customizable.
Drawings
Fig. 1 is a schematic structural diagram of an intention-driven cloud access network system provided in an embodiment of the present invention;
in the figure: 1. a service application layer; 2. an intent-to-enable layer; 3. an infrastructure layer.
Fig. 2 is a flowchart of an intention-driven clouding access network method provided in an embodiment of the present invention.
FIG. 3 is a functional diagram of an intent engine provided by an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The invention aims at diversified task requirements in the next five generations of mobile communication, realizes resource sharing and dynamic scheduling by an intention-driven cloud ubiquitous access network and adopting a software defined network and network function virtualization technology, and improves spectrum efficiency so as to achieve low-cost, high-bandwidth and flexible network operation.
The following detailed description of the principles of the invention is provided in connection with the accompanying drawings.
As shown in fig. 1, an intention-driven cloud access network system provided in an embodiment of the present invention includes: business application layer 1, intent enable layer 2, infrastructure layer 3.
The service application layer 1 is used for programming the bottom layer equipment through a programming interface provided by the intention enabling layer, abstractly packaging the capability of the network element and opening the control right of the network to users and other service partners; and a management interface is provided, and colorful business innovation is realized.
The intention enabling layer 2 is used for analyzing and checking consistency of an intention flow translated by the application request through the intention northbound interface and processing the intention into a network intention which can be executed by the current network resource; and obtaining the corresponding operation of the specific resource in the network through the mapping algorithm of the intention and the resource. The system based on the management and arrangement of the intents is adopted to realize the unified planning of the global application intents, the unified scheduling of resources and the closed-loop arrangement mode to realize the life cycle management of the network elements.
And the infrastructure layer 3 is used for independently providing non-real-time functions CU-C and CU-U of one or more access modes to form an access aggregation layer and carrying out unified cooperative management on DUs of one or more access modes (such as NR and eLTE). Meanwhile, the enhanced radio remote unit is deployed, so that the problems of fast change of network conditions, large service demand, complex user behavior attributes and insufficient coverage are solved.
As shown in fig. 2, the method for an intention-driven cloud access network according to an embodiment of the present invention includes the following steps:
s201: translation of application intent: the intention northbound interface embodies the application intention through a declarative expression rule and deconstructs the intention into a specific expression form of < action, object, modification, type. >; the translated intention flow enters an intention-based management and orchestration system (IB-MANO) of a cloud network service plane through a REST interface;
s202: analysis of intent and consistency check: an intention analysis module of the intention stream entering the intention engine carries out the request of identifying the application intention and carries out the rule matching on the operation expression in the intention stream; based on a commitment theory, an intention consistency checking module is used for solving the problem of a large amount of logic consistency caused by possible intention conflict when network delay occurs or a plurality of application programs send intents at the same time;
s203: the strategy mapping obtains a resource combination scheme: aiming at the analyzed atomic request of the intention flow, determining the resource combination by combining with the resource combination rule stored in the resource combination template library; analyzing the priority and conflict of the resource combination scheme, and determining the type of the virtual network element;
s204: resource configuration parameter calculation: the parameter calculation module generates configuration parameters according to the working load and the environmental information and by combining the historical parameter decision information learned in the self-learning module; the determined resource composition and parameters are generated in YAML format and transmitted to the optimization verification module;
s205: and (3) continuous closed loop optimization verification: and comparing the obtained configuration parameters with the real-time network state and the historical data in the optimization verification module to verify the correctness of the parameter configuration. Finally, the data file is delivered to a virtualized network function manager to complete the arrangement and management; after the application intents are realized, the optimization verification module continuously verifies the reliability of the intention realization and realizes data updating in cooperation with the self-learning module;
s206: obtaining feedback information of an access network side: and the centralized management node (CU) establishes a network service integral data model by adopting technologies such as big data, machine learning and the like, transmits key index data to the self-learning module and the optimization verification module through a feedback link in real time, realizes scene prejudgment at the CU and carries out pre-allocation scheduling on resources.
In the preferred embodiment of the invention, the conflict and priority analysis firstly determines the network configuration attribute associated with the executed policy action according to the input policy event and the priority relation matrix; if for each network configuration attribute, a plurality of policies related to the network configuration attribute are adopted at the same time, a conflict occurs; according to the priority information defined in the priority matrix, the network configuration attribute associated with the policy with the highest priority is executed, and other policies are refused to be executed. The priority relation matrix is constructed by taking a network element as a unit, and the matrix describes the priority of the strategy behavior adopted to operate the relevant configuration attributes under the condition that the judgment condition of the strategy is met.
In a preferred embodiment of the present invention, the self-learning module employs a supervised learning mechanism, i.e. a linear regression model and a deep neural network, to implement the self-learning mechanism. In addition, the self-learning module defines an upper limit of the processing time in advance to ensure the timely realization of the application intention.
The application of the principles of the present invention will now be described in further detail with reference to the accompanying drawings.
As shown in FIG. 3, the intent engine of the present invention includes seven modules, an intent parsing module to understand and learn the intent flow translated via the northbound interface; the intention consistency checking module is used for solving the problem that a great amount of logic consistency problems are generated due to possible intention conflicts when network delay occurs or a plurality of application programs send intents simultaneously, and the resource mapping module realizes automatic acquisition of resource combination decisions through rule matching; the resource combination template stores the resource combination template and defines the basic resource composition of each basic service; the parameter calculation module determines a virtual network element configuration parameter according to the service requirement and the resource constraint; the self-learning module adopts a parameter decision scheme model constructed by supervised learning based on empirical data;
the optimization verification module is used for pre-verifying the validity and the correctness of the configuration scheme and simultaneously enabling the network to obtain the self-optimization and self-healing capabilities in the long-term feedback verification.
The method for generating the resource action flow from the intention flow by the intention engine provided by the embodiment of the invention comprises the following steps:
step one, analyzing intention and checking consistency.
(1) The intention analysis module divides a group of data of the intention flow into parts of speech and loads labels, and realizes rule matching for each labeled symbol. Outputting the analyzed intention request in a JSON file; an intent flow may be expressed as < action, object, modifier, type. >, an action is the most critical operation describing an intent, an object identifies the actor or recipient of the action, a modifier is used to materialize or parameterize a certain constraint, and a type is an identification of a connection to another service process.
(2) Based on a commitment theory, an intention consistency checking module is used for solving the problem of a large amount of logic consistency caused by possible intention conflict when network delay occurs or a plurality of application programs send intents at the same time; when the intention reaches, different intention targets are analyzed, an optimized solution is generated to meet the requirement, and after the conflict is solved at the upper layer (control layer), the solution is issued to the lower layer to execute the intention without conflict. The form of the commitment theory can be expressed as:
wherein n is1Is a commitment person, n2Is the recipient of the commitment, and pi is the type of commitment, describing the commitment characteristics. The type of commitment can be divided into basic, cooperation, use and condition.
In a network, the commitment taker and the recipient of the commitment may be various nodes in the network, with the commitment theory characterizing the relationship between pairs of nodes. The intent should be to satisfy all the set of operations that involve commitments between nodes or to maximally satisfy the commitments of all the nodes involved.
And step two, obtaining a resource combination scheme by strategy mapping.
(1) Aiming at the analyzed atomic request of the intention flow, determining the resource combination by combining with the resource combination rule stored in the resource combination template library;
(2) and analyzing the priority and conflict of the resource combination scheme, determining the type of the virtual network element, and analyzing the service level, the service reliability and the service safety.
And step three, calculating resource configuration parameters.
(1) And the parameter calculation module generates configuration parameters according to the working load and the environmental information and by combining the historical parameter decision information learned in the self-learning module. The self-learning module adopts a supervised learning mechanism, namely a linear regression model and a deep neural network to realize the self-learning mechanism. In addition, the self-learning module defines an upper limit of the processing time in advance to ensure the timely realization of the application intention.
(2) The determined resource composition and parameters are generated in YAML format and passed to the optimization verification module.
And step four, continuously performing closed-loop optimization verification.
(1) And comparing the obtained configuration parameters with the real-time network state and the historical data in the optimization verification module to verify the correctness of the parameter configuration. And finally, delivering the data file to a virtual network function manager to complete the arrangement and management.
(2) After the intention is realized, the optimization verification module continuously verifies the reliability of the intention realization and realizes data updating in cooperation with the self-learning module.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (7)
1. An intent-driven clouded access network system, characterized in that the intent-driven clouded access network system comprises:
the service application layer programs the bottom layer equipment through a programming interface provided by the intention enabling layer, abstractly encapsulates the capability of the network element, and opens the control right of the network to users and other service partners; and providing a management interface;
the intention enabling layer is used for analyzing and checking consistency of an intention flow translated by the application request through the intention northbound interface and processing the intention into a network intention capable of being executed by the current network resource; obtaining corresponding operation of specific resources in the network through a mapping algorithm of the intents and the resources; the system based on the management and arrangement of the intentions is adopted to realize the unified planning of the global application intentions, the unified scheduling of resources and the closed-loop arrangement mode to realize the life cycle management of the network elements;
and the infrastructure layer is used for independently providing the non-real-time functions CU-C and CU-U of one or more access systems to form an access aggregation layer and carrying out unified cooperative management on the DUs of one or more access systems.
2. The intent-driven clouded access network system of claim 1, wherein the intent-driven clouded access network system further comprises an intent engine module, the intent engine module comprising:
an intent parsing module for understanding and learning intent streams translated via the northbound interface;
the system comprises an intention checking module, a logic consistency checking module and a logic consistency checking module, wherein the intention checking module is used for solving the problem of a large amount of logic consistency caused by possible intention conflicts when network delay occurs or a plurality of application programs send intents at the same time;
the resource mapping module realizes the automatic acquisition of resource combination decision through rule matching;
the resource combination template library is used for storing resource combination templates and defining the basic resource composition of each basic service;
the parameter calculation module is used for determining the configuration parameters of the virtual network element according to the service requirements and the resource constraints;
the autonomous learning module is used for constructing a parameter decision scheme model by adopting supervised learning based on empirical data;
and the optimization verification module is used for pre-verifying the validity and the correctness of the configuration scheme and simultaneously enabling the network to obtain the self-optimization and self-healing capabilities in the long-term feedback verification.
3. An intention-driven clouding access network method applying the intention-driven clouding access network system described in claim 1, characterized in that the intention-driven clouding access network method comprises the steps of:
step one, an intention northbound interface embodies an application intention through an declarative expression rule and deconstructs the application intention into a specific expression form of actions, objects, modification and types; the translated intention flow enters an intention-based management and organization system IB-MANO of a cloud network service plane through an intention northbound interface;
step two, the intention flow enters an intention analysis module in an intention engine to identify a request in the application intention, and rule matching is carried out on an operation indicator in the intention flow; based on a commitment theory, an intention consistency checking module is used for solving the problem of a large amount of logic consistency caused by possible intention conflict when network delay occurs or a plurality of application programs send intents at the same time;
step three, aiming at the analyzed request of the intention flow, determining the resource combination by combining the resource combination rule stored in the resource combination template library; analyzing the priority and conflict of the resource combination scheme, and determining the type of the virtual network element;
fourthly, the parameter calculation module generates configuration parameters according to the working load and the environmental information and by combining the historical parameter decision information learned in the self-learning module; the determined resource composition and parameters are generated in YAML format and transmitted to the optimization verification module;
comparing the obtained configuration parameters with the real-time network state and historical data in an optimization verification module to verify the correctness of parameter configuration; the data file is delivered to a virtual network function manager to complete the arrangement and management; after the application intents are realized, the optimization verification module continuously verifies the reliability of the intention realization and realizes data updating in cooperation with the self-learning module;
step six, access network side feedback information acquisition: the centralized management node establishes a network service integral data model by adopting big data and machine learning technology, transmits key index data to the self-learning module and the optimization verification module through a feedback link in real time, realizes scene prejudgment at the management node and carries out pre-allocation scheduling of resources.
4. The intent-driven clouded access network method of claim 3, wherein said conflict and priority analysis specifically comprises:
(1) determining the network configuration attribute associated with the executed strategy behavior according to the input strategy event and the priority relation matrix; if for each network configuration attribute, a plurality of policies related to the network configuration attribute are adopted at the same time, a conflict occurs;
(2) according to the priority information defined in the priority matrix, executing the network configuration attribute associated with the policy with the highest priority, and refusing to execute other policies;
(3) the priority relation matrix is constructed by taking a network element as a unit, and the matrix describes the priority of the strategy behavior adopted to operate the relevant configuration attributes under the condition that the judgment condition of the strategy is met.
5. The intent-driven clouding access network method of claim 3, wherein the self-learning module employs a supervised learning mechanism, a linear regression model and a deep neural network to implement the self-learning mechanism; the self-learning module defines an upper processing time limit in advance to ensure timely implementation of the application intention.
6. An intention engine implementing generation method from an intention flow to a resource action flow by applying the intention-driven clouded access network method of claim 3, wherein the intention engine implementing generation method from the intention flow to the resource action flow comprises:
step one, analyzing intention and checking consistency;
(1) the intention analysis module divides a group of data of the intention stream into parts of speech and loads labels, and realizes rule matching for each label attached symbol; outputting the analyzed intention request in a JSON file; intent flow is represented as an action, an object, a modifier, a type, an action is the most critical operation describing an intent, an object-is the executor or acceptor identifying an action, a modifier is used to materialize or parameterize a constraint, a type is the identification of a connection to another service process;
(2) based on the commitment theory, analyzing different intention targets when the intention arrives, generating an optimized solution to meet the requirement, and issuing the solution to a lower layer to execute a conflict-free intention after the conflict is solved at an upper layer; the form of the commitment theory is expressed as:
wherein n is1Is a commitment person, n2Is the recipient of the commitment, pi is the type of commitment, describes the commitment characteristics; the type of commitment is divided into basis, cooperation, use and condition;
in the network, the commitment device and the commitment receiver can be various nodes in the network, and the relationship between the node pairs is characterized through the commitment theory; the set of operations intended to satisfy all commitments involving nodes or to maximally satisfy all node commitments involved;
step two, strategy mapping is carried out to obtain a resource combination scheme;
(1) the analyzed atomic request of the intention flow is combined with the resource combination rule stored in the resource combination template library to realize the determination of the resource combination;
(2) analyzing the priority and conflict of the resource combination scheme, determining the type of the virtual network element, and analyzing the service level, the service reliability and the service safety;
step three, calculating resource configuration parameters;
(1) the parameter calculation module generates configuration parameters according to the working load and the environmental information and by combining the historical parameter decision information learned in the self-learning module; the self-learning module adopts a supervised learning mechanism, namely a linear regression model and a deep neural network to realize the self-learning mechanism;
(2) the determined resource composition and parameters are generated in YAML format and transmitted to the optimization verification module;
step four, continuously performing closed-loop optimization verification;
(1) comparing the obtained configuration parameters with the real-time network state and the historical data in an optimization verification module to verify the correctness of the parameter configuration; the data file is delivered to a virtual network function manager to complete the arrangement and management;
(2) after the intention is realized, the optimization verification module continuously verifies the reliability of the intention realization and realizes data updating in cooperation with the self-learning module.
7. A wireless communication network system applying the intention-driven cloud access network method of any one of claims 3 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810929724.2A CN109245916B (en) | 2018-08-15 | 2018-08-15 | Intention-driven cloud access network system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810929724.2A CN109245916B (en) | 2018-08-15 | 2018-08-15 | Intention-driven cloud access network system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109245916A CN109245916A (en) | 2019-01-18 |
| CN109245916B true CN109245916B (en) | 2021-06-11 |
Family
ID=65070384
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810929724.2A Active CN109245916B (en) | 2018-08-15 | 2018-08-15 | Intention-driven cloud access network system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109245916B (en) |
Families Citing this family (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115051903B (en) * | 2019-02-14 | 2023-08-04 | 华为技术有限公司 | Intent processing method, device and system |
| CN110198237B (en) * | 2019-05-27 | 2021-05-04 | 北京邮电大学 | Configuration method of wireless intention driven network |
| CN110278111B (en) * | 2019-05-29 | 2021-08-31 | 西安电子科技大学 | General architecture of intention-driven network and translation method of intention-driven network |
| CN110247795B (en) * | 2019-05-30 | 2020-09-25 | 北京邮电大学 | Intent-based cloud network resource service chain arranging method and system |
| CN110166305B (en) * | 2019-06-04 | 2020-04-21 | 电子科技大学 | Method and system for managing application QoE (quality of experience) based on intention |
| CN111193603B (en) * | 2019-08-06 | 2021-10-22 | 腾讯科技（深圳）有限公司 | Network architecture generation method and device, readable storage medium and computer equipment |
| CN110868315B (en) * | 2019-09-23 | 2022-11-01 | 中移（杭州）信息技术有限公司 | Information processing method of northbound interface |
| CN110673860B (en) * | 2019-09-24 | 2023-06-16 | 南京先进计算产业发展有限公司 | Intelligent cloud data center construction and operation deployment system and method |
| CN112583629B (en) * | 2019-09-30 | 2022-06-10 | 华为技术有限公司 | Information processing method, related equipment and computer storage medium |
| CN112738723B (en) * | 2019-10-11 | 2023-04-07 | 中国电信股份有限公司 | Network resource allocation method and device and computer readable storage medium |
| CN111210126A (en) * | 2019-12-27 | 2020-05-29 | 中国电力科学研究院有限公司 | Application program system and method for ubiquitous connectable electric energy system |
| CN113825128B (en) * | 2020-06-20 | 2022-12-30 | 华为技术有限公司 | Communication method, device and system |
| CN112631856B (en) * | 2020-12-17 | 2022-04-29 | 西安电子科技大学 | Intelligent operation and maintenance method, system, equipment and application of intention-driven 6G network |
| CN112953778B (en) * | 2021-03-17 | 2023-04-18 | 北京交通大学 | Intention-driven-based service arrangement system and method in intelligent fusion identification network |
| CN114167760B (en) * | 2021-10-23 | 2023-12-29 | 西安电子科技大学 | Intention driven network management system and method |
| CN114137861A (en) * | 2021-10-23 | 2022-03-04 | 西安电子科技大学 | Intention-driven cloud security service system and method |
| CN114172814B (en) * | 2021-10-23 | 2023-02-07 | 西安电子科技大学 | Method for constructing intention-driven satellite network resource management triple-cooperation model and application |
| CN116155737A (en) * | 2021-11-23 | 2023-05-23 | 华为云计算技术有限公司 | Network configuration method and device |
| CN114666330B (en) * | 2022-02-15 | 2024-02-06 | 浪潮通信信息系统有限公司 | Method, device, equipment and product for arranging calculation network |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106708997A (en) * | 2016-12-20 | 2017-05-24 | 浪潮通信信息系统有限公司 | Requirement-driven Web service aggregation method in business domain framework |
| CN106713244A (en) * | 2015-11-17 | 2017-05-24 | 中国移动通信集团公司 | Capability access method and network element |
-
2018
- 2018-08-15 CN CN201810929724.2A patent/CN109245916B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106713244A (en) * | 2015-11-17 | 2017-05-24 | 中国移动通信集团公司 | Capability access method and network element |
| CN106708997A (en) * | 2016-12-20 | 2017-05-24 | 浪潮通信信息系统有限公司 | Requirement-driven Web service aggregation method in business domain framework |
Non-Patent Citations (1)
| Title |
|---|
| Intent-based management and orchestration of heterogeneous openflow/IoT SDN domains;Walter Cerroni 等;《2017 IEEE Conference on Network Softwarization (NetSoft)》;20170707;第1-9页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109245916A (en) | 2019-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109245916B (en) | Intention-driven cloud access network system and method | |
| CN112953778B (en) | Intention-driven-based service arrangement system and method in intelligent fusion identification network | |
| Saeik et al. | Task offloading in Edge and Cloud Computing: A survey on mathematical, artificial intelligence and control theory solutions | |
| Zhang et al. | Cyber-physical-social systems: The state of the art and perspectives | |
| CN110278111B (en) | General architecture of intention-driven network and translation method of intention-driven network | |
| US10262019B1 (en) | Distributed management optimization for IoT deployments | |
| CN114167760B (en) | Intention driven network management system and method | |
| Boudi et al. | AI-based resource management in beyond 5G cloud native environment | |
| EP4293965A1 (en) | Information processing method, apparatus, system, electronic device and storage medium | |
| Zeb et al. | Industry 5.0 is coming: A survey on intelligent nextG wireless networks as technological enablers | |
| Guim et al. | Autonomous lifecycle management for resource-efficient workload orchestration for green edge computing | |
| US11463322B1 (en) | Digital twin architecture for internet-connected device computing environment | |
| CN114143181B (en) | Intention-driven spatial information network arrangement system and method | |
| Okwuibe et al. | Sdn-enabled resource orchestration for industrial iot in collaborative edge-cloud networks | |
| US10922623B2 (en) | Capacity planning, management, and engineering automation platform | |
| CN114143272A (en) | Intention-driven 6G ground network management and control system and method | |
| Kafle et al. | Network control and management automation: Architecture standardization perspective | |
| Guimarães et al. | Deep: A vertical-oriented intelligent and automated platform for the edge and fog | |
| Zhang et al. | The intelligent multi‐domain service function chain deployment: Architecture, challenges and solutions | |
| Zhou et al. | Knowledge transfer and reuse: A case study of AI-enabled resource management in RAN slicing | |
| TW202031067A (en) | System for wifi route discovery in virtual network | |
| Verginadis et al. | PrEstoCloud: Proactive Cloud Resources Management at the Edge for Efficient Real-Time Big Data Processing. | |
| Zeb et al. | Towards defining industry 5.0 vision with intelligent and softwarized wireless network architectures and services: A survey | |
| Wang et al. | A new approach to heterogeneous wireless sensor networks reliability evaluation based on perception layer in internet of vehicles | |
| Benmerar et al. | Intelligent Multi-Domain Edge Orchestration for Highly Distributed Immersive Services: An Immersive Virtual Touring Use Case |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |