AI-Powered Network Resource Management System Based on Dynamic Carbon Awareness

Abstract

This invention relates to an AI-powered network resource management system for reducing carbon emissions in 5G, 5G-Advanced, and 6G communication networks. The invention evaluates the instantaneous carbon density of the power grid together with the renewable energy production and energy storage status of network nodes, and calculates a dynamic carbon suitability score for each candidate path or resource cluster via an AI decision engine. Traffic routing, bandwidth allocation, and network resource management operations are performed based on this calculated carbon suitability score. Thanks to this invention, the total network carbon footprint and energy consumption are reduced, the use of renewable energy is increased, and service quality is maintained.

Description

1 TARIFF AI-Powered Network Resource Management Based on Dynamic Carbon Awareness The system Technical Area 5 The invention addresses energy and carbon footprint in next-generation communication networks such as 5G, 5G-Advanced, and 6G. It relates to network resource management systems aimed at increasing efficiency. More specifically, the invention measures the instantaneous carbon density of the electricity grid, at the point where the network nodes are located. renewable energy production status, availability of energy storage systems, grid load and by evaluating service quality requirements together, an AI-powered dynamic network 10 It relates to a system that performs resource management. The invention relates to core networks, access networks (RAN), IP-based transport networks, and MPLS. It can be used in SDN and in future next-generation communication infrastructures. State of the Art 15 Today, network resource management in communication networks mostly involves quality of service, latency, This is done by taking parameters such as bandwidth and traffic load into account. There are studies aimed at energy efficiency and reducing carbon emissions. However, the vast majority of current systems only consider the energy consumption of network nodes. taking into account the instantaneous carbon intensity of the electricity grid and the renewable energy of network nodes 20 It does not consider energy production and storage situations together. In particular, some network nodes in new generation communication networks use solar or wind energy. While some nodes are supported by energy storage systems, others are directly powered by electricity. It operates depending on the grid. The carbon density of the electricity grid over time These 25 can vary and are included in current network resource management systems. It does not take changes into account. In addition, current methods may affect future carbon costs or the impact on the network. It does not include dynamic decision-making mechanisms aimed at reducing the total carbon footprint. In conclusion, due to the drawbacks described above and the limitations of the current solutions, Due to its inadequacy, it has become necessary to make improvements in the relevant technical field. 30 It has arrived. Purpose of the Invention The invention was created by drawing inspiration from existing techniques and overcoming the aforementioned drawbacks. It aims to eliminate. 35 2 The main purpose of the invention is to determine the instantaneous carbon density of the electrical grid and the network nodes. Artificial intelligence by evaluating renewable energy production and energy storage situations together. The goal is to provide a system that implements supported dynamic network resource management. Another aim of the invention is to provide, for each candidate path or resource set on the network;  Carbon intensity score, 5  energy consumption score,  renewable energy eligibility score,  energy storage suitability score,  network load score and  Quality of service (QoS) score 10 The goal is to create a multi-criteria decision-making mechanism through calculation. Another aim of the invention is to determine the instantaneous carbon density of the electrical grid and the network nodes. By evaluating renewable energy production amounts together, the total network carbon footprint to reduce. Another aim of the invention is to create renewable energy sources with high or low carbon intensity. The goal is to ensure that network nodes located in those regions are used prioritized. Another purpose of the invention is to manage traffic routing, capacity allocation, and network nodes on the network. activating or deactivating and allocating resources based on carbon awareness. to accomplish it as follows. Another aim of the invention is to reduce total energy consumption and carbon 20% while maintaining service quality. The goal is to reduce emissions while simultaneously increasing the use of renewable energy. Another objective of the invention is to provide a performance monitoring and feedback mechanism. an autonomous network resource management system capable of adapting to changing network conditions to create. To achieve the objectives described above, the invention contributes to dynamic carbon awareness. It relates to an AI-powered network resource management system. Figures that will help understand the invention. Figure 1 shows the general architecture of the system that is the subject of the invention. Explanation of Part References 1. Network data acquisition module 2. Energy consumption calculation module 3. Carbon intensity determination module 4. Renewable energy assessment module 35 5. Energy storage evaluation module 3 6. Network load and service quality evaluation module 7. Artificial intelligence decision engine 8. Carbon compliance score calculation module 9. Resource selection module 10. Network resource management module 5 11. Performance monitoring and feedback module Detailed Description of the Invention In this detailed explanation, the preferred configurations of the invention are only a better understanding of the subject. This information is provided for the purpose of understanding and will not have any limiting effect. 10 The invention relates the instantaneous carbon density of the electricity grid to the renewable energy of network nodes. AI-powered dynamic system by evaluating production and energy storage conditions together. AI-powered, dynamic carbon awareness-based network resource management system. It is related to network resource management systems. In an example application of the system that is the subject of the invention shown in Figure 1, network data collection 15 module (1) provides information about network topology, node and link loads, traffic demands, energy It collects consumption data, service quality parameters, and network status information. Energy consumption calculation module (2), obtained by network data collection module (1) Using this information, it calculates the current and historical energy consumption of network nodes. Carbon intensity determination module (3), instantaneous carbon intensity of the electricity grid, energy 20 Carbon output for each network node using production sources and carbon emission data. It determines the density information. Renewable energy assessment module (4), solar energy, wind energy of network nodes or the amount of energy they obtain from similar renewable energy sources and It evaluates their usability. 25 Energy storage evaluation module (5), batteries or similar located in network nodes utilization rates and availability of energy storage systems and energy storage It determines their capabilities. Network load and service quality assessment module (6) assesses traffic load, latency, on the network. It analyzes bandwidth usage, packet loss, and quality of service parameters. 30 Artificial intelligence decision engine (7), energy consumption calculation module (2), carbon intensity determination module (3), renewable energy assessment module (4), energy storage by the evaluation module (5) and the network load and quality of service evaluation module (6) Using the information generated, dynamic carbon suitability for each candidate route or resource set. Calculates the score. 35 4 In a sample application of the invention, the dynamic carbon compliance score is as follows: is being calculated: 𝐶𝑆 = 𝑤 𝐺𝐶𝐼 + 𝑤 𝑅𝐸𝑆 + 𝑤 𝐵𝐴𝑇 + 𝑤 𝐸 +𝑤 𝑄 +𝑤 𝐿 Here are 5  𝐶𝑆: Dynamic carbon compliance score,  𝐺𝐶𝐼: Carbon intensity score of the electricity grid,  𝑅𝐸𝑆: Renewable energy eligibility score,  𝐵𝐴𝑇: Energy storage suitability score,  E: Energy consumption score, 10  𝑄: Service quality score,  𝐿: Network load score,  𝑤 ,𝑤 ,𝑤 ,𝑤 ,𝑤 : weighting coefficients of the relevant parameters It expresses. The carbon compliance score calculation module (8) is obtained by the artificial intelligence decision engine (7) 15 Using the results obtained, carbon suitability assessments were calculated for each candidate route or resource set. It produces. The source selection module (9) evaluates carbon compliance values ​​and service quality. from among the candidate routes or resource sets that meet the requirements, the most suitable one chooses. 20 Network resource management module (10), selects the routing path or resource set for the network It implements this on the network. This includes traffic routing, bandwidth allocation, and network node assignment. Network resource management operations such as activating or putting the network into sleep mode can be performed. Performance monitoring and feedback module (11) monitors energy consumption, carbon on the network. emissions, renewable energy use rate, service quality and network performance continuously 25 It monitors as follows. The information obtained is fed back to the artificial intelligence decision engine (7) and the system It allows the parameters to be updated. This invention reduces the network's total carbon footprint compared to existing network resource management methods. by reducing our digital footprint, increasing the use of renewable energy, decreasing energy consumption, and It is able to maintain service quality. 30

Claims

REQUESTS 1. Instantaneous carbon intensity of the electricity grid and renewable energy production of network nodes. and energy storage conditions are evaluated together by an AI-powered dynamic network. Dynamic Carbon Awareness-Based Artificial Intelligence that performs resource management It is a Supported Network Resource Management System, and its feature is; 5 • Network topology, traffic load, connection status, power consumption, and network health information. Network data collection module (1), • energy consumption calculator that calculates the instantaneous and historical energy consumption of network nodes. calculation module (2), • instantaneous carbon intensity of the electricity grid, energy production sources and carbon 10 Determining carbon intensity information for each network node using emission data. carbon intensity determination module (3), • network nodes powered by solar, wind, or similar renewable energy sources the amount of energy they obtain from their sources and their availability renewable energy assessment module (4), 15 • batteries or similar energy storage systems located in network nodes determining their occupancy rates, availability, and energy storage capacities energy storage evaluation module (5), • network traffic load, latency, bandwidth usage, packet loss, and service Network load and service quality assessment module 20, which analyzes quality parameters. (6), • energy consumption, carbon intensity, renewable energy production, energy storage By evaluating the situation, network load, and service quality information together, each candidate will be guided through the process. artificial intelligence that calculates a dynamic carbon compliance score for a source set or resource group decision engine (7), 25 • Using the results obtained by the artificial intelligence decision engine (7), each candidate path or carbon suitability that calculates carbon suitability values ​​for a resource set score calculation module (8), • Candidates that meet service quality requirements according to carbon compliance values Resource selection module 30, which selects the most suitable option from among paths or resource sets. (9), • implements the selected routing path or resource set on the network, traffic Routing, bandwidth allocation, and enabling or disabling network nodes in sleep mode. Network resource management module (10) capable of performing the transfer operations, • energy consumption, carbon emissions, renewable energy usage rate, service quality 35 and continuously monitors network performance and uses the information obtained to make artificial intelligence decisions. 6 by feeding it back to the motor (7) enabling the system parameters to be updated Performance monitoring and feedback module (11) It includes.

2. According to claim 1, it is a network resource management system, and its feature is; an artificial intelligence decision engine (7) carbon intensity of the electricity grid, renewable energy production status, energy 5 storage status, energy consumption, network load, and quality of service parameters together. by evaluating each candidate route or resource set with a dynamic carbon compliance score. It is a calculation.

3. Network resource management system according to claim 1 or 2, with the feature of dynamic carbon compliance 10 the carbon intensity score of the electricity grid, renewable energy eligibility score, energy storage availability score, energy consumption score, quality of service score, and network load The score is calculated using a weighted decision function.

4. A network resource management system according to any of claims 1-3, and its characteristic is; renewable. energy evaluation module (4) solar energy, wind energy, hydroelectric energy 15 or using data obtained from similar renewable energy sources and energy storage evaluation module (5) battery or similar energy storage It is the evaluation of data related to the systems.

5. Network resource management system according to any of claims 1-4, and its characteristic is performance. Monitoring and feedback module (11), energy consumption, carbon emissions, renewable 20 by monitoring energy usage rate, latency, packet loss and service quality parameters is updating the parameters of the artificial intelligence decision engine (7).

Images