Digital twin enriched green topology system for next generation core networks and operation method thereof

Abstract

The invention relates to a digital twin enriched green topology system for next generation core networks, aided by a computer with at least one processor, and the operation method thereof. More specifically, the invention relates to a green topology system that enables topology discovery, one of the key functions of next generation core network management, to be performed with less resource (CPU) usage and less energy consumption, and the operation method thereof.

Description

[0001] DIGITAL TWIN ENRICHED GREEN TOPOLOGY SYSTEM FOR NEXT GENERATION CORE NETWORKS AND OPERATION METHOD THEREOF

[0002] Technical Field of the Invention

[0003] The invention relates to a digital twin enriched green topology system for next generation core networks, aided by a computer with at least one processor, and the operation method thereof.

[0004] State of the Art

[0005] Over the past few years, due to rising energy costs and the critical environmental impact of network devices, energy efficiency in core network discovery has become a key consideration for Internet Service Providers (ISPs). At this point, intelligent and resource- aware discovery approaches are crucial to provide reliable network management, taking into account the convergence time of the network. However, traditional discovery does not take into account energy consumption issues and requires a high amount of resources, such as the Central Processing Unit (CPU). In addition, due to the lack of predictive methods, traditional discovery cannot provide rapid convergence time.

[0006] Topology discovery is one of the key functions of core network management, as it enables the mapping of active network devices at the physical and logical layers, but it has complexities in terms of operational and resource efficiency. To give an example, traditional discovery fails to perform predictive analysis to learn network behavior, and moreover, traditional discovery periodically visits IP ports without considering the usage levels of router ports, leading to high resource usage and energy consumption. Therefore, it is necessary to integrate intelligent methods into traditional discovery to deeply understand the behavioral pattern of a core network and avoid these complexities.

[0007] In the present art, there are significant initiatives to integrate Digital Twin (DT) technology into network management and control systems to improve communication services. For example, the activation of loT applications over sixth-generation (6G) networks to manage and optimize 6G edge networks is contemplated. In addition, 6G Digital Twin Networking (DTN) is contemplate from both data and communication perspectives by establishing a reference architecture for interfaces between layers to form a closed control loop. However, none of these studies addressed a problem focused on network discovery.

[0008] In the state of the art, the existing literature on router visit behavior is divided into two parts, TPD (Traditional Partial Discovery) and TCD (Traditional Complete Discovery), from the perspective of network discovery. In TPD, instead of periodic visits, a subset of the topology is selected and visited; however, its usage in the telecommunications sector has not yet become widespread. Existing TPDs and TCDs are able to provide accurate topology information for different network types (sensor networks, software-defined networks, and drone networks); however, only the studies for sensor networks address the issue of energy consumption.

[0009] It has become necessary to present a system and an operation method thereof that overcomes all the problems mentioned in the present art and enables the discovery of core networks with less resource and energy consumption.

[0010] Summary and Objects of the Invention

[0011] The invention describes a digital twin enriched green topology system for next generation core networks, aided by a computer with at least one processor, and the operation method thereof. More specifically, the invention describes a green topology system that enables topology discovery, one of the key functions of next generation core network management, to be performed with less resource (CPU) usage and less energy consumption, and the operation method thereof. In the inventive system, high- performance virtualization of Digital Twin (DT) technology and the high-accuracy prediction capability of Machine Learning (ML) are utilized. Within the scope of the invention, a digital twin layer of the physical IP core is created to sustain more efficient discovery management where descriptive, predictive analytics are enabled to learn network behavior and perform state analysis. In this layer, the application of the energy consumption formula specially derived for the discovery service is carried out. In addition, the invention uses a Multilayer Perceptron (MLP) algorithm to decide which routers to perform port-level discovery on. The object of the invention is to enable the discovery of core networks with reduced resource usage and energy consumption. The inventive system uses Digital twin-based energy-friendly discovery approach (DT-GDP) to enable the discovery of core networks with reduced resource usage and energy consumption. In this new approach, a new formula is used for the energy consumption of network discovery, and the information of the ports to be visited for discovery is decided based on machine learning. In this context, the outputs of the two modules are used to calculate the total energy consumption in Watts. In the energy module, the power consumption when an IP (Internet Protocol) port is actively serving, the power consumption in the idle state and the power consumption when the cooling is active are taken into account. In the visit decision module, Multilayer Perceptron (MLP) is used to classify IP ports and suggest a visit action. The main reasons for using DT technology in the new discovery approach used in the inventive system are that;

[0012] • DT enables real-time remote monitoring and control throughout the entire lifecycle of the discovery service, which allows predictive and prescriptive analytics to be deployed and probabilistic analytics to be run

[0013] • Unlike traditional discovery services, DT is able to support services with a direct connection to network operational data instead of a recently updated IP inventory. The recently updated IP inventory becomes obsolete as time goes on until the next discovery service cycle begins. Therefore, exploration integrated with DT enables more accurate topology information to be obtained through intelligent observation and control.

[0014] • Unlike traditional discovery services, DT (Digital Twin) supports the application of various Machine Learning methods to predict unnecessary visits and thus reduce costs.

[0015] Description of the Drawings

[0016] Figure 1. Representative illustration of an inventive digital twin enriched green topology system for next-generation core networks, aided by a computer with at least one processor. Figure 2. Representative illustration of the operation method of an inventive digital twin enriched green topology system for next-generation core networks, aided by a computer with at least one processor.

[0017] Description of the References in Figures

[0018] 1 . Physical layer module

[0019] 2. Virtual layer module

[0020] 3. Virtual model

[0021] 4. Energy module

[0022] 5. Discovery decision module

[0023] 6. Discovery component module

[0024] 1001. Receiving the data collected from the Physical Layer module (1 ) by the virtual layer module (2)

[0025] 1002. Creating virtual models (3) with the data received by the virtual layer module (2)

[0026] 1003. Evaluating the number of virtual models (3) in the environment and if there are at least 2 active virtual models, proceeding to energy calculation and discovery decision processes, and if there are no at least two active models, continuing data collection from the physical layer module (1 )

[0027] 1004. Creating a formula for the power consumption of network discovery in the energy module (4) using the virtual model data, and in this formula, using the power consumption information noted when the discovery service is running on the router, the power consumption information noted when no service is running on this router (idle state), and the cooling power consumption information required when the router is active

[0028] 1005. Estimating which routers and ports will be discovered with the help of Multilayer Perceptron algorithm in the discovery decision module (5) using virtual model data

[0029] 1006. Evaluating the outputs of the energy module (4) and the discovery decision module (5) together and running the network discovery algorithm that minimizes energy consumption Detailed Description of the Invention

[0030] The invention relates to a digital twin enriched green topology system for next generation core networks, aided by a computer with at least one processor, and the operation method thereof. More specifically, the invention relates to a green topology system that enables topology discovery, one of the key functions of next generation core network management, to be performed with less resource (CPU) usage and less energy consumption, and the operation method thereof. An inventive digital twin enriched green topology system for next-generation core networks, aided by a computer with at least one processor comprises;

[0031] • a physical layer module (1 ) representing the core network where the internet service provided by internet service providers (ISPs) to their users is offered, and containing routers of the P (provider) and PE (provider edge) types,

[0032] • a virtual layer module (2), which is the layer in which the virtual twins of each of the routers in the physical layer module (1 ) and the modules that enable intelligent network discovery are located,

[0033] • a virtual model (3), which is the copies in the virtual layer of the physical routers on the core network, holding all the operational and configurational data of the physical router,

[0034] • energy module (4), which is the module to which the power consumption formula specially derived for the network discovery service is applied,

[0035] • a discovery decision module (5), which is the module in which machine learning- supported estimation of the router and related ports to be visited in topology discovery is made,

[0036] • the discovery component module (6), which is the module where the outputs of the Energy Module and the Discovery Decision Module are evaluated together and the algorithm that performs the discovery to all routers and ports that need to be discovered by minimizing energy consumption is implemented.

[0037] The operation method of an inventive digital twin enriched green topology system for next-generation core networks, aided by a computer with at least one processor comprises the process steps of; i. receiving the data collected from the physical layer module (1 ) by the virtual layer module (2) (1001 ), ii. creating virtual models (3) with the data received by the virtual layer module (2) (1002), iii. evaluating the number of virtual models (3) in the environment and if there are at least 2 active virtual models, proceeding to energy calculation and discovery decision processes, and if there are no at least two active models, continuing data collection from the physical layer module (1 ) (1003), iv. creating a formula for the power consumption of network discovery in the energy module (4) using the virtual model data, and in this formula, using the power consumption information noted when the discovery service is running on the router, the power consumption information noted when no service is running on this router (idle state), and the cooling power consumption information required when the router is active (1004), v. estimating which routers and ports will be discovered with the help of Multilayer Perceptron algorithm in the discovery decision module (5) using virtual model data (1005), vi. evaluating the outputs of the energy module (4) and the discovery decision module (5) together and running the network discovery algorithm that minimizes energy consumption (1006).

[0038] The Digital Twin supported network discovery mentioned in the inventive system essentially consists of two layers, the physical layer module (1 ) and the virtual layer module (2). In the physical layer module (1 ), there are physical routers that make up today's core network structure. These are categorized as P (provider) and PE (provider edge) according to their types. The data collected from said P and PE routers in the physical layer module (1 ) are transmitted to the virtual layer module (2). In the virtual layer module (2), virtual models (3) of the routers in the physical layer module (1 ) are created. Each virtual model has all the qualities of the physical router it expresses. After the creation of virtual models (3) is completed, the number of active virtual models is evaluated. If this number is less than the value of "2", the algorithm re-establishes the data flow from the physical layer module (1 ) to the virtual layer module (2) and increases the number of virtual models (3) in the environment. If this number is at a value of at least 2, the energy calculation discovery estimation process continues. In this context, the data collected from virtual models are processed in the energy module (4) and a special energy consumption formula is created for the network discovery service. In this formula, the power consumption information noted when the discovery service is running on the router, the power consumption information noted when no service is running on this router (idle state), and the cooling power consumption information required when the router is active are used. Likewise, the data collected from virtual models (3) is also processed in the discovery decision module (5) to decide which routers and ports to perform discovery on. Multilayer Perceptron (MLP) machine learning method is utilized in this decision mechanism. Finally, by evaluating the outputs of the energy module (4) and the discovery decision module (5) together, the discovery algorithm is run in the discovery component module (6) to keep the energy consumption level of access to all routers and ports to be discovered to a minimum.

[0039] The power consumption formula of the network discovery service derived under energy module (4) is as follows;

[0040] 1p udle As a result of the Multilayer Perceptron (MLP) algorithm applied within the scope of the discovery decision module (5), the information of the routers and ports to be visited is decided as follows;

[0041] Formula 2.

[0042] The algorithm run in Discovery Component (6) calculates the total power consumption value given below:

[0043] Formula 3.

Claims

CLAIMS1. A digital twin enriched green topology system for next generation core networks, aided by a computer with at least one processor, characterized in that it comprises;• a physical layer module (1 ) representing the core network where the internet service provided by internet service providers (ISPs) to their users is offered, and containing routers of the P (provider) and PE (provider edge) types,• a virtual layer module (2), which is the layer in which the virtual twins of each of the routers in the physical layer module (1 ) and the modules that enable intelligent network discovery are located,• a virtual model (3), which is the copies in the virtual layer of the physical routers on the core network, holding all the operational and configurational data of the physical router,• energy module (4), which is the module to which the power consumption formula specially derived for the network discovery service is applied,• a discovery decision module (5), which is the module in which machine learning-supported estimation of the router and related ports to be visited in topology discovery is made,• the discovery component module (6), which is the module where the outputs of the Energy Module and the Discovery Decision Module are evaluated together and the algorithm that performs the discovery to all routers and ports that need to be discovered by minimizing energy consumption is implemented.

2. The operation method of a digital twin enriched green topology system for next generation core networks, aided by a computer with at least one processor, characterized in that it comprises the process steps of; i. receiving the data collected from the physical layer module (1 ) by the virtual layer module (2) (1001 ), ii. creating virtual models (3) with the data received by the virtual layer module (2) (1002), iii. evaluating the number of virtual models (3) in the environment and if there are at least 2 active virtual models, proceeding to energy calculation and discovery decision processes, and if there are no at least two active models, continuing data collection from the physical layer module (1 ) (1003), iv. creating a formula for the power consumption of network discovery in the energy module (4) using the virtual model data, and in this formula, using the power consumption information noted when the discovery service is running on the router, the power consumption information noted when no service is running on this router (idle state), and the cooling power consumption information required when the router is active (1004), v. estimating which routers and ports will be discovered with the help of Multilayer Perceptron algorithm in the discovery decision module (5) using virtual model data (1005), vi. evaluating the outputs of the energy module (4) and the discovery decision module (5) together and running the network discovery algorithm that minimizes energy consumption (1006).