Method and system for managing computing power session based on 6G communication, sensing and computing integration network

Abstract

The invention discloses a network management computing power session method and system based on 6G communication, sensing and computing integration, belongs to the technical field of mobile communication, and aims to solve the technical problem of how to realize flexible scheduling and on-demand distribution of computing power and network convergence resources in a 6G communication, sensing and computing integration network environment under the condition of not changing network equipment. A mobile network environment supporting a computing power-as-a-service mode is constructed for a user, and the adopted technical scheme is as follows: the method comprises the following steps of: deploying a computing network integration sensing network element, a computing network integration control scheduling network element and a computing network integration service arrangement network element on a core network side, so that the improved core network has a communication, sensing and computing power integration function; and through analysis and identification of computing power application request indexes, computing power business arrangement, computing power and network Qos configuration mapping, decision making and scheduling of computing network fusion resources, distributed cooperative computing tasks are issued, so that computing power session creation is completed, and different computing power requirements of the mobile terminal are met.

Description

Technical Field

[0001] The present invention relates to the field of mobile communication technology, and in particular to a method and system for managing computing power conversation based on a 6G integrated network. Background Art

[0002] At present, 5G network session management mainly focuses on ensuring network service quality and ensuring data transmission reliability, low latency and high bandwidth. However, when facing emerging AI computing applications, 5G network session management has the limitation of not being able to directly perceive the computing power requirements on the terminal side and not being able to allocate computing power and network resources to the terminal side at the same time. This is mainly because the original design and core functions of 5G networks focus more on improving communication capabilities rather than directly supporting the allocation and management of computing resources.

[0003] In 6G networks, the deep integration of communication, perception and computing capabilities to form an integrated network is one of the important directions of 6G development. However, in the design and development of 6G networks, although the deep integration of communication, perception and computing capabilities is emphasized, there is a lack of specific protocols, standards and mature technical solutions in the specific implementation of session and computing application task binding strategies and session management computing power and network integration indicators.

[0004] Therefore, in the 6G integrated network environment, how to achieve flexible scheduling and on-demand allocation of computing power and network fusion resources without changing network equipment, and build a mobile network environment that supports the computing power as a service model for users is a technical problem that needs to be solved urgently. Summary of the invention

[0005] The technical task of the present invention is to provide a method and system for managing computing power sessions based on a 6G integrated computing network, so as to solve the problem of how to achieve flexible scheduling and on-demand allocation of computing power and network fusion resources in a 6G integrated computing network environment without changing network equipment, and to build a mobile network environment that supports the computing power as a service model for users.

[0006] The technical task of the present invention is achieved in the following way: a method for managing computing power sessions based on a 6G integrated network of communication and computing. The method deploys computing-network fusion perception network elements, computing-network fusion control and scheduling network elements, and computing-network fusion service orchestration network elements on the core network side, so that the improved core network has communication, perception and computing power integration functions; and completes computing power session creation by analyzing and identifying computing power application request indicators, computing power service orchestration, computing power and network QoS configuration mapping, as well as decision-making and scheduling computing-network fusion resources, and issuing distributed collaborative computing tasks to meet the different computing power requirements of mobile terminals.

[0007] As a preferred option, the computing-network fusion perception network element detects and collects surrounding network quality status and computing power resource information in real time and builds a computing power perception model and a network perception model;

[0008] The computing-network converged service orchestration network element is used to identify service requests and execute policies for service orchestration;

[0009] The computing-network fusion control and scheduling network element is used to convert policies into specific parameters and perform resource scheduling.

[0010] Preferably, after the computing-network fusion perception network element is started, it collects network quality status information and computing power resource information of surrounding devices in real time, and builds corresponding network perception models and computing power perception models based on the network quality perception information and computing power resource information, respectively, to form a dynamic computing power and network resource pool; among which, the network quality status information includes bandwidth, delay, jitter, packet loss rate and load information; the computing power resource information includes computing power device type, computing power measurement value, computing power storage performance and size information.

[0011] Preferably, the decision-making and scheduling computing network integration resources are as follows:

[0012] After the mobile terminal accesses the improved core network, it initiates a computing power session creation request to the service-oriented interface of the computing network fusion service orchestration network element through the HTTP2 protocol, carrying the computing service type and the required network quality requirements and inference computing power requirement fields.

[0013] Preferably, the analysis and identification of computing power application request indicators and computing power and network QoS configuration mapping are as follows:

[0014] The computing-network converged service orchestration network element analyzes and identifies the service indicators of the computing power session request;

[0015] The results of analysis and identification are used to calculate the total business computing power network demand measurement value through data modeling. The formula is as follows:

[0016]

[0017] Where E represents the total business computing power network demand measurement value; f c Indicates the computing power requirements of the business; f n Indicates the computing power requirements of the business in terms of network; α and β are the weight ratio coefficients corresponding to computing and network;

[0018] Through the table lookup matching method, the QoS (Quality of Service) model index with the highest similarity to the total business computing power network demand measurement value is retrieved from the local configuration, so as to convert the business demand index into a computing power QoS configuration list and a network QoS configuration list; among them, the computing power QoS configuration list includes computing coordination mode, number of resource nodes, computing priority and load scheduling; the network QoS configuration list includes network service category, maximum uplink and downlink rate, minimum guaranteed rate and preemption reservation mechanism.

[0019] Preferably, the computing power service orchestration is as follows:

[0020] When a mobile terminal is connected to the improved core network, it initiates the creation of a computing session based on the AI ​​computing application. The computing-network fusion service orchestration network element selects a service orchestration strategy based on the network quality requirements and computing requirements requested by the computing application and the information collected by the computing-network fusion perception network element. It uses data modeling methods to comprehensively consider the network status, computing node load and service priority, and achieve full utilization and optimal configuration of computing resources.

[0021] Distributed collaborative computing tasks are generated according to the business orchestration strategy, and a dedicated computing channel is created for the session. The specific task information is sent to the selected computing power node, thereby completing the allocation of computing resources and opening up the data transmission path, so that end-user applications can obtain good computing and network integration services through computing power sessions.

[0022] Preferably, the distributed collaborative computing task is issued to establish a computing power session as follows:

[0023] After receiving the network QoS configuration, the computing network fusion control and scheduling network element converts the configuration of the QoS configuration list into network parameters, and sends it to the policy control network element through the existing N5 interface of the core network, creating a computing bearer data channel designed specifically for computing power sessions between the terminal and the site, and between the base station and the core network;

[0024] After receiving the computing power QoS configuration list, the computing network fusion control and scheduling network element converts the configuration in the computing power QoS configuration list into computing power parameters, establishes computing tasks for the computing power session, selects computing collaboration methods and computing nodes, as well as candidate computing nodes and computing path information, and then sends specific task information to the selected computing nodes to pre-allocate computing power and network resources and establish computing data plane transmission tunnels; among them, computing nodes include master control nodes and slave node clusters.

[0025] Preferably, after receiving the response from the computing-network convergence service orchestration network element indicating successful session creation, the mobile terminal obtains the main control node (usually the terminal itself) in the response information, and the main control node starts the computing task and notifies the slave nodes to start task collaborative computing.

[0026] A 6G-based integrated network management computing power session system, which deploys computing network fusion perception network elements, computing network fusion control and scheduling network elements, and computing network fusion service orchestration network elements on the core network side, so that the improved core network has communication, perception and computing power integration functions;

[0027] Among them, the computing network fusion perception network element is used to detect and collect the surrounding network quality status and computing power resource information in real time and build computing power perception model and network perception model;

[0028] The computing-network converged service orchestration network element is used to identify service requests and execute policies for service orchestration;

[0029] The computing-network fusion control and scheduling network element is used to convert policies into specific parameters and perform resource scheduling.

[0030] Preferably, the system further includes a main control node, a wireless base station and a computing power server; the working process of the system is as follows:

[0031] (0) The main control node, wireless base station and computing power server register computing power with the computing network fusion perception network element respectively;

[0032] (1) The main control node initiates a computing power session creation request to the computing network convergence service orchestration network element;

[0033] (2) The computing-network converged service orchestration network element analyzes the service request indicators and maps the indicators into a computing power QoS configuration list and a network QoS configuration list;

[0034] (3) The computing network convergence service orchestration NE sends the service orchestration results to the computing network convergence control and scheduling NE;

[0035] (4) The computing-network fusion control and scheduling network element converts the network QoS configuration into network parameters and creates a computing bearer channel for the session on the core network side;

[0036] (5) The computing-network fusion control and scheduling network element sends specific tasks to computing servers, wireless base stations, and main control nodes; the specific task information includes task type, coordination mode, node information, and transmission tunnel information;

[0037] (6) The computing network fusion control scheduling network element responds to the successful creation of the computing power session, obtains the main control node in the response information, and the main control node starts the computing task and notifies the wireless base station and computing power server to start the task collaborative computing.

[0038] The 6G synergy-computing integrated network management computing power session method and system of the present invention have the following advantages:

[0039] (1) The present invention is applied to computing task type session requests initiated by mobile terminals based on 6G networks, and solves the problem of computing task session requests initiated by the terminal side in a 6G integrated network environment. By adding self-developed software and protocols, flexible scheduling and on-demand allocation of computing power and network fusion resources can be achieved without changing network equipment, thereby building a mobile network environment that supports the "computing power as a service" (CaaS) model for users;

[0040] (ii) The present invention solves the allocation and scheduling of computing power and network resources for computing services in mobile sessions under the 6G inter-sensory computing network environment, fully calls on the effective computing power resources of edge-end-cloud devices to form collaborative distributed computing, which not only provides a stable network transmission environment with large bandwidth, reliability and low latency, but also greatly reduces the computing time of services through the optimized configuration of computing power resources;

[0041] (III) The present invention is applied under the 6G integrated network architecture. On the basis of retaining the original network architecture and functions, by adding computing-network fusion perception network elements, computing-network fusion control and scheduling network elements, and computing-network fusion service orchestration network elements on the core network side, the network is not only responsible for communication, but also has the ability of resource perception and unified scheduling of computing and networks; on this basis, the session management function is newly added with the process of parsing and identifying computing power application request indicators, the process of mapping computing power and network QoS configuration, the process of computing power service orchestration, the process of decision-making and scheduling computing-network fusion resources, and the process of issuing distributed collaborative computing tasks, so that ordinary sessions are transformed into computing power sessions to meet the different computing power requirements of terminals;

[0042] (IV) The present invention integrates network quality requirements and computing power requirements into session request indicators. The QoS configuration mapping also associates computing power and network indicators at the same time. In subsequent orchestration and scheduling, computing power and network resources are integrated and used together, so that the session can obtain the support of computing power resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention is further described below in conjunction with the accompanying drawings.

[0044] Attached Figure 1 This is the network architecture diagram of 6G integrated sensing and computing;

[0045] Attached Figure 2 This is a flow chart of the 6G Tongsensing integrated network management computing power conversation system;

[0046] Attached Figure 3 This is a schematic diagram of the computing network fusion QoS model. DETAILED DESCRIPTION

[0047] The following detailed description is made of the network management computing power session method and system based on 6G integrated communication computing with reference to the drawings and specific embodiments of the specification.

[0048] Embodiment 1:

[0049] As attached Figure 1 As shown, this embodiment provides a method for managing computing power sessions based on a 6G inter-sensing computing integration network. The method is to deploy computing network fusion perception network elements, computing network fusion control scheduling network elements, and computing network fusion service orchestration network elements on the core network side, so that the improved core network has communication, perception and computing power integration functions; and complete the creation of computing power sessions by parsing and identifying computing power application request indicators, computing power service orchestration, computing power and network QoS configuration mapping, as well as decision-making and scheduling computing network fusion resources, and issuing distributed collaborative computing tasks to meet the different computing power requirements of mobile terminals; when the mobile terminal is connected to the improved core network, the mobile terminal initiates a computing power application based on AI computing power. Computing power session creation, computing-network fusion service orchestration network element selects service orchestration strategy based on network quality requirements and computing power requirements requested by computing power application and the results of information collection by computing-network fusion perception network element. It uses data modeling method to comprehensively consider network status, computing power node load machine service priority to achieve full utilization and optimal configuration of computing power resources. Finally, it produces distributed collaborative computing tasks according to the service orchestration strategy, creates a dedicated computing channel for the session, and sends specific task information to the selected computing power node, thereby completing the allocation of computing resources and opening up the data transmission path, so that end-user applications can obtain good computing-network fusion services through computing power sessions.

[0050] The computing-network fusion perception network element in this embodiment detects and collects surrounding network quality status and computing power resource information in real time and builds a computing power perception model and a network perception model.

[0051] The computing-network convergence service orchestration network element in this embodiment is used to identify service requests and execute policies for service orchestration.

[0052] The computing-network fusion control and scheduling network element in this embodiment is used to convert policies into specific parameters and perform resource scheduling.

[0053] The specific steps of this embodiment are as follows:

[0054] S1. After the computing-network fusion perception network element is started, the network quality status information and computing power resource information of the surrounding devices are collected in real time, and the corresponding network perception model and computing power perception model are constructed based on the network quality perception information and the computing power resource information respectively, to form a dynamic computing power and network resource pool; among which, the network quality status information includes bandwidth, delay, jitter, packet loss rate and load information; the computing power resource information includes the computing power device type, computing power measurement value, computing power storage performance and size information;

[0055] S2. After the mobile terminal accesses the improved core network, it initiates a computing power session creation request to the service-oriented interface of the computing network fusion service orchestration network element through the HTTP2 protocol, carrying the computing service type and the required network quality requirements and inference computing power requirement fields.

[0056] S3. The computing-network fusion service orchestration network element parses and identifies the service indicators of the computing power session request, and calculates the total service computing power network demand metric value through data modeling. The formula is as follows:

[0057]

[0058] Where E represents the total business computing power network demand measurement value; f c Indicates the computing power requirements of the business; f n Indicates the computing power demand of the business in terms of network; α and β are the weight ratio coefficients corresponding to computing and network; then, through the table matching method, the QoS (Quality of Service) model index with the highest similarity to the total business computing power network demand metric value is retrieved from the local configuration, as shown in the attached Figure 3 As shown, the business demand indicators are converted into a computing power QoS configuration list and a network QoS configuration list; the computing power QoS configuration list includes computing coordination mode, number of resource nodes, computing priority and load scheduling; the network QoS configuration list includes network service category, maximum uplink and downlink rate, minimum guaranteed rate and preemption reservation mechanism;

[0059] S4. After receiving the network QoS configuration, the computing network integration control and scheduling network element converts the configuration of the QoS configuration list into network parameters, and sends it to the policy control network element through the existing N5 interface of the core network, creating a computing bearer data channel designed specifically for computing power sessions between the terminal and the site, and between the base station and the core network;

[0060] S5. After receiving the computing power QoS configuration list, the computing network fusion control scheduling network element converts the configuration in the computing power QoS configuration list into computing power parameters, establishes computing tasks for the computing power session, selects computing collaboration mode and computing nodes, as well as candidate computing nodes and computing path information, and then sends specific task information to the selected computing nodes to pre-allocate computing power and network resources and establish computing data plane transmission tunnels; wherein, the computing nodes include master control nodes and slave node clusters;

[0061] S6. After the mobile terminal receives the response from the computing-network convergence service orchestration network element indicating that the session is successfully created, it obtains the main control node (usually the terminal itself) in the response information. The main control node starts the computing task and notifies the slave nodes to start task collaborative computing.

[0062] Embodiment 2:

[0063] This embodiment provides a 6G communication-sensing-computing integrated network management computing power session system, which deploys computing-network fusion perception network elements, computing-network fusion control and scheduling network elements, and computing-network fusion service orchestration network elements on the core network side, so that the improved core network has communication, perception, and computing power integration functions;

[0064] Among them, the computing network fusion perception network element is used to detect and collect the surrounding network quality status and computing power resource information in real time and build computing power perception model and network perception model;

[0065] The computing-network converged service orchestration network element is used to identify service requests and execute policies for service orchestration;

[0066] The computing-network fusion control and scheduling network element is used to convert policies into specific parameters and perform resource scheduling.

[0067] As attached Figure 2 As shown in the figure, the working process of the system is as follows:

[0068] (0) The mobile terminal device, wireless base station and computing power server register computing power with the computing network fusion perception network element respectively;

[0069] (1) The mobile terminal device initiates a computing session creation request to the computing network convergence service orchestration network element;

[0070] (2) The computing-network converged service orchestration network element analyzes the service request indicators and maps the indicators into a computing power QoS configuration list and a network QoS configuration list;

[0071] (3) The computing network convergence service orchestration NE sends the service orchestration results to the computing network convergence control and scheduling NE;

[0072] (4) The computing-network fusion control and scheduling network element converts the network QoS configuration into network parameters and creates a computing bearer channel for the session on the core network side;

[0073] (5) The computing-network fusion control and scheduling network element sends specific tasks to computing servers, wireless base stations, and mobile terminal devices; the specific task information includes task type, coordination mode, node information, and transmission tunnel information;

[0074] (6) The computing network fusion control scheduling network element responds to the successful creation of the computing power session, obtains the main control node in the response information, and the main control node starts the computing task and notifies the wireless base station and computing power server to start the task collaborative computing.

[0075] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for managing computing power session based on 6G integrated network management, characterized in that: The method is to deploy computing-network fusion perception network elements, computing-network fusion control and scheduling network elements, and computing-network fusion service orchestration network elements on the core network side, so that the improved core network has integrated communication, perception, and computing power functions; It also completes computing session creation by analyzing and identifying computing application request indicators, computing business orchestration, computing and network QoS configuration mapping, decision-making and scheduling of computing and network fusion resources, and issuing distributed collaborative computing tasks to meet the different computing power requirements of mobile terminals.

2. The method for managing computing power session based on 6G integrated network management based on claim 1 is characterized in that: The computing-network fusion perception network element detects and collects surrounding network quality and computing resource information in real time and builds computing power perception models and network perception models; The computing-network converged service orchestration network element is used to identify service requests and execute policies for service orchestration; The computing-network fusion control and scheduling network element is used to convert policies into specific parameters and perform resource scheduling.

3. The method for managing computing power session based on 6G integrated network management based on claim 1 or 2, characterized in that: After the computing-network fusion perception network element is started, it collects network quality status information and computing power resource information of surrounding devices in real time, and builds corresponding network perception models and computing power perception models based on network quality perception information and computing power resource information respectively, to form a dynamic computing power and network resource pool; among them, network quality status information includes bandwidth, delay, jitter, packet loss rate and load information; computing power resource information includes computing power device type, computing power measurement value, computing power storage performance and size information.

4. The method for managing computing power session based on 6G integrated network management based on claim 3 is characterized in that: The decision-making and scheduling computing network integration resources are as follows: After the mobile terminal accesses the improved core network, it initiates a computing power session creation request to the service-oriented interface of the computing network fusion service orchestration network element through the HTTP2 protocol, carrying the computing service type and the required network quality requirements and inference computing power requirement fields.

5. The method for managing computing power session based on 6G integrated network management based on claim 4 is characterized in that: The specific details of parsing and identifying computing power application request indicators and computing power and network QoS configuration mapping are as follows: The computing-network converged service orchestration network element analyzes and identifies the service indicators of the computing power session request; The results of analysis and identification are used to calculate the total business computing power network demand measurement value through data modeling. The formula is as follows: Where E represents the total business computing power network demand measurement value; f c Indicates the computing power requirements of the business; f n Indicates the computing power requirements of the business in terms of network; α and β are the weight ratio coefficients corresponding to computing and network; Through the table lookup matching method, the QoS model index with the highest similarity to the total business computing power network demand measurement value is retrieved from the local configuration, so as to convert the business demand index into a computing power QoS configuration list and a network QoS configuration list; among them, the computing power QoS configuration list includes computing coordination mode, number of resource nodes, computing priority and load scheduling; the network QoS configuration list includes network service category, maximum uplink and downlink rate, minimum guaranteed rate and preemption reservation mechanism.

6. The method for managing computing power session based on 6G integrated network management based on claim 5 is characterized in that: The specific arrangement of computing power business is as follows: When a mobile terminal is connected to the improved core network, it initiates the creation of a computing session based on the AI ​​computing application. The computing-network fusion service orchestration network element selects a service orchestration strategy based on the network quality requirements and computing requirements requested by the computing application and the information collected by the computing-network fusion perception network element. It uses data modeling methods to comprehensively consider the network status, computing node load and service priority, and achieve full utilization and optimal configuration of computing resources. Distributed collaborative computing tasks are generated according to the business orchestration strategy, and a dedicated computing channel is created for the session. The specific task information is sent to the selected computing power node, thereby completing the allocation of computing resources and opening up the data transmission path, so that end-user applications can obtain good computing and network integration services through computing power sessions.

7. The method for managing computing power session based on 6G integrated network management based on claim 6 is characterized in that: The details of issuing distributed collaborative computing tasks to establish a computing power session are as follows: After receiving the network QoS configuration, the computing network fusion control and scheduling network element converts the configuration of the QoS configuration list into network parameters, and sends it to the policy control network element through the existing N5 interface of the core network, creating a computing bearer data channel designed specifically for computing power sessions between the terminal and the site, and between the base station and the core network; After receiving the computing power QoS configuration list, the computing network fusion control and scheduling network element converts the configuration in the computing power QoS configuration list into computing power parameters, establishes computing tasks for the computing power session, selects computing collaboration methods and computing nodes, as well as candidate computing nodes and computing path information, and then sends specific task information to the selected computing nodes to pre-allocate computing power and network resources and establish computing data plane transmission tunnels; among them, computing nodes include master control nodes and slave node clusters.

8. The method for managing computing power session based on 6G integrated network management based on claim 7 is characterized in that: After the mobile terminal receives the response from the computing-network convergence service orchestration network element that the session is successfully created, it obtains the main control node in the response information, the main control node starts the computing task, and notifies the slave nodes to start task collaborative computing.

9. A 6G-based integrated network management computing power conversation system, characterized in that: The system deploys computing-network fusion perception network elements, computing-network fusion control and scheduling network elements, and computing-network fusion service orchestration network elements on the core network side, so that the improved core network has integrated communication, perception, and computing power functions; Among them, the computing network fusion perception network element is used to detect and collect the surrounding network quality status and computing power resource information in real time and build computing power perception model and network perception model; The computing-network converged service orchestration network element is used to identify service requests and execute policies for service orchestration; The computing-network fusion control and scheduling network element is used to convert policies into specific parameters and perform resource scheduling.

10. The method for managing computing power session based on 6G integrated network management based on claim 9 is characterized in that: The working process of the system is as follows: (0) The mobile terminal device, wireless base station and computing power server register computing power with the computing network fusion perception network element respectively; (1) The mobile terminal device initiates a computing session creation request to the computing network convergence service orchestration network element; (2) The computing-network converged service orchestration network element analyzes the service request indicators and maps the indicators into a computing power QoS configuration list and a network QoS configuration list; (3) The computing network convergence service orchestration NE sends the service orchestration results to the computing network convergence control and scheduling NE; (4) The computing-network fusion control and scheduling network element converts the network QoS configuration into network parameters and creates a computing bearer channel for the session on the core network side; (5) The computing-network fusion control and scheduling network element sends specific tasks to computing servers, wireless base stations, and mobile terminal devices; the specific task information includes task type, coordination mode, node information, and transmission tunnel information; (6) The computing network fusion control scheduling network element responds to the successful creation of the computing power session, obtains the main control node in the response information, and the main control node starts the computing task and notifies the wireless base station and computing power server to start the task collaborative computing.

Images