An advanced artificial intelligence-based traffic analysis system for digital twin networks and working method thereof

Abstract

The invention is related to a system that aims to increase network performance and security by performing artificial intelligence-based traffic analysis within Digital Twin Networks (DTN) and the working method of this system.

Description

[0001] DESCRIPTION

[0002] AN ADVANCED ARTIFICIAL INTELLIGENCE-BASED TRAFFIC ANALYSIS SYSTEM FOR DIGITAL TWIN NETWORKS AND WORKING METHOD THEREOF

[0003] Technical Field

[0004] The invention is related to a system that aims to increase network performance and security by performing artificial intelligence-based traffic analysis within Digital Twin Networks (DTN) and the working method of this system. The invention aims to provide high quality and real-time data collection, the use of scalable and efficient artificial intelligence models, ensuring explainability and transparency, security and privacy protection, real-time processing and energy efficiency.

[0005] State of the Art

[0006] Many applications and algorithms are used today for network management and performance optimization. Traditional network management systems often rely on static configurations and predefined rules, and struggle to adapt to both dynamic and complex network environments. These systems are known to be insufficient for realtime performance optimization and anomaly detection.

[0007] Some networks use basic Machine Learning (ML) or Deep Learning (DL) models for traffic prediction or simple anomaly detection. These models are often trained with limited data sets, and in addition, their real-time processing capability is limited. In addition, it has shortcomings in scalability and adaptation to complex network topologies.

[0008] A digital twin is defined as a real-time virtual model of an object or system. A digital twin uses real data about a real-life object or system as input and generates predictions and control strategies about how the real object or system will react. A network digital twin, on the other hand, is a virtual representation of a communication network that accurately models the devices, communication links, operating environment, and applications running on the network. There are existing applications for the use of artificial intelligence techniques in digital twin networks, however these applications often involve artificial intelligence integration at limited scope or for specific tasks. Thelack of explainability and transparency of artificial intelligence models leads to trust issues in decision-making processes, and vital vulnerabilities can occur when adequate measures are not taken on data security and privacy.

[0009] One of the technical problems that arise as a result of these shortcomings is the inability to effectively perform real-time, scalable, and secure traffic analysis in digital twin networks. Current applications are unable to provide real-time and scalable traffic analysis in the complex and dynamic environment of digital twin networks. Data quality and accessibility issues reduce the performance of artificial intelligence models and lead to wrong decisions. Security and privacy concerns reduce the reliability and adoption of artificial intelligence-based systems.

[0010] Therefore, it is necessary to develop a method that minimizes or eliminates the above-mentioned disadvantages in the state of the art and a system that works according to said method.

[0011] Summary of the Invention

[0012] One of the advantages of the invention is the ability to collect high quality and real-time data to solve the problems identified in the present art, enabling the use of scalable and efficient artificial intelligence models, ensuring explainability and transparency, security and privacy protection, real-time processing and energy efficiency.

[0013] Another advantage of the invention is that users and administrators can be informed by using explainable artificial intelligence models to provide explainability and transparency. Thus, decision-making processes are made more understandable. Security and privacy protection are provided, and artificial intelligence-based anomaly detection and protection mechanisms against cyber-attacks are integrated.

[0014] Another advantage of the invention is that for energy efficiency and resource optimization, resource allocation will be optimized with optimization algorithms, thus reducing energy consumption. With modular and flexible system architecture for integration and compatibility, easy integration and expandability options are offered.

[0015] Description of the DrawingsFigure 1: A representative schematic illustration of an inventive advanced artificial intelligence-based traffic analysis system for digital twin networks.

[0016] Figure 2: A representative flow diagram illustration of the general working method of an inventive advanced artificial intelligence-based traffic analysis system for digital twin networks.

[0017] Figure 3: A representative flow diagram illustration showing the steps to be taken to maintain the integrity of the collected data in the working method of an inventive advanced artificial intelligence-based traffic analysis system for digital twin networks. Figure 4: A representative flow diagram illustration showing the working method of an advanced artificial intelligence-based traffic analysis system for digital twin networks, showing a flow of the user-oriented development of the system.

[0018] Figure 5: It is a representative flow diagram illustration showing the working method of an inventive advanced artificial intelligence-based traffic analysis system for digital twin networks, showing the steps to increase the energy efficiency of the network and improve resource utilization.

[0019] Description of the References in the Drawings

[0020] For a better understanding of the invention, the description of the numbers in the figures is given below:

[0021] 100. System

[0022] 1. Data collection module

[0023] 2. Data cleaning and pre-processing module

[0024] 3. Simulation and modeling module

[0025] 4. Analysis module

[0026] 5. Prediction module

[0027] 6. Detection module

[0028] 7. Decision-making module

[0029] 8. Response module

[0030] 9. Artificial intelligence model

[0031] 10. Communication interface

[0032] 11. Security and privacy module

[0033] 12. User interface and reporting module13. Virtual network

[0034] 14. Physical network

[0035] 15. Optimization algorithms

[0036] 1001. Collecting raw data from physical networks (14) with at least one data collection module (1)

[0037] 1002. Cleaning and pre-processing collected data with at least one data cleaning and pre-processing module (2)

[0038] 1003. Using the pre-processed data to create virtual models with at least one simulation and modeling module (3)

[0039] 1004. Analyzing data with various artificial intelligence models (9) through at least one analysis module (4)

[0040] 1005. Predicting future network trends and potential problems with at least one prediction module (5)

[0041] 1006. Detecting anomalies and security threats in the network with at least one detection module (6)

[0042] 1007. Making decisions about network optimization according to the results of the analysis by using optimization algorithms (15) with at least one decision-making module (7)

[0043] 1008. Implementing decisions made with at least one response module (8) to the physical network (14) via the communication interface (10)

[0044] 1009. Ensuring data security and privacy with at least one security and privacy module (11)

[0045] 1010. Presenting reports and visualizations to users with at least one user interface and reporting module (12)

[0046] 2001. Securely collecting and transmitting data with at least one data collection module (1)

[0047] 2002. Ensuring data integrity using blockchain technologies and Federative Learning (FL) artificial intelligence model with at least one security and privacy module (11) 2003. Using the pre-processed data to create virtual models with at least one simulation and modeling module (3)

[0048] 2004. Analyzing data using artificial intelligence models (9) with at least one analysis module (4)3001. Collecting data on system performance and status with at least one data collection module (1)

[0049] 3002. Creating reports and visualizations with at least one user interface and reporting module (12)

[0050] 3003. Presenting reports to users and administrators with at least one user interface and reporting module (12)

[0051] 3004. Gathering and evaluating user feedback with at least one user interface and reporting module (12)

[0052] 3005. Improving the system based on feedback with optimization algorithms (15)

[0053] 4001. Selecting and implementing optimization algorithms

[0054] 4002. Optimizing resource allocation and energy efficiency

[0055] 4003. Determining performance improvement strategies

[0056] 4004. Evaluating optimization results and adjustment if necessary

[0057] Detailed Description of the Invention

[0058] The example embodiments are described in more detail below with reference to the accompanying descriptions. Furthermore, the embodiments can be established in different forms and should not be interpreted as being limited to the embodiments specified herein. Rather, these example embodiments are provided so that this description will be thorough, and will fully convey the scope to those skilled in the art.

[0059] The terminology used in this description is intended to describe a particular example embodiment only and is not intended to be limiting. As used herein, the forms "a", "at least", and "preferably" are intended to include plural forms as well, unless the context clearly states otherwise.

[0060] The invention relates to an advanced artificial intelligence-based traffic analysis system for digital twin networks and the working method of this system. A system (100) performing artificial intelligence-based traffic analysis with a computer comprising at least one processor comprises; at least one data collection module (1) that collects raw data, at least one data cleaning and pre-processing module (2) that cleans and performs pre-processing steps of the collected data, at least one simulation and modeling module (3) that creates models and simulations using the pre-processeddata, at least one analysis module (4) that performs analysis with various artificial intelligence techniques, at least one prediction module (5) that predicts future network trends and potential problems, at least one detection module (6) that detects anomalies and potential security threats, at least one decision-making module (7) that makes decisions for network optimization based on the analysis results, at least one response module (8) that applies the decision made to the physical network (14), at least one artificial intelligence model (9), at least one communication interface (10) that ensures data and commands flow between modules, at least one security and privacy module (11) that ensures data security and privacy, at least one user interface and reporting module (12) that provides users and administrators with reports and visualizations on the status of the system, its performance, and decisions made, at least one virtual network (13) which comprises virtual models of physical networks (14), at least one physical network (14) which is the source of the data and where the optimizations are applied, and optimization algorithms (15) which comprise algorithms for performance improvement and are used in the decision-making module (7).

[0061] The working method an inventive advanced artificial intelligence-based traffic analysis system for digital twin comprises the process steps of:

[0062] I. Collecting raw data from physical networks (14) with at least one data collection module (1) (1001),

[0063] II. Cleaning and pre-processing collected data with at least one data cleaning and pre-processing module (2) (1002),

[0064] III. Using the pre-processed data to create virtual models with at least one simulation and modeling module (3) (1003),

[0065] IV. Analyzing data with various artificial intelligence models (9) through at least one analysis module (4) (1004),

[0066] V. Predicting future network trends and potential problems with at least one prediction module (5) (1005),

[0067] VI. Detecting anomalies and security threats in the network with at least one detection module (6) (1006),

[0068] VII. Making decisions about network optimization according to the results of the analysis by using optimization algorithms (15) with at least one decision-making module (7) (1007),VIII. Implementing decisions made with at least one response module (8) to the physical network (14) via the communication interface (10) (1008),

[0069] IX. Ensuring data security and privacy with at least one security and privacy module (11) (1009),

[0070] X. Presenting reports and visualizations to users with at least one user interface and reporting module (12) (1010).

[0071] The invention is a system that aims to increase network performance and security by performing artificial intelligence-based traffic analysis within Digital Twin Networks (DTN). The starting point of the inventive system (100) is physical networks (14), representing real-world devices and network structure. Thanks to the data collection module (1), raw data is collected from these networks. The collected data is processed with the data cleaning and pre-processing module (2) to increase the data quality and reliability, thus making the data suitable for analysis.

[0072] With the processed data, a representation of virtual networks (13) is created through the simulation and modeling module (3). These virtual networks act as digital twins of physical networks and serve as the basis for simulation processes.

[0073] The analysis module (4) analyzes the processed data in depth with artificial intelligence models (9). In this part, artificial intelligence models such as Machine Learning (ML), Deep Learning (DL), Reinforcement Learning (RL), Federative Learning (FL), and Graph Neural Networks (GNN) are used. Thanks to these models, important information about performance and network status is obtained.

[0074] The results of the analysis are used by the prediction module (5) and the detection module (6). The Prediction Module predicts future network trends and potential problems and enables measures to be taken for the future. The detection module (6) detects anomalies and potential security threats in the network in real-time.

[0075] Thanks to the information obtained, the decision-making module (7) plays a decisionmaking role for network-specific optimization using optimization algorithms (15). The decisions made are applied to the physical network (14) via the communication interface (10) with the response module (8). This process takes place with automatic and real-time response mechanisms, completing the feedback loop this way.Data security and privacy are ensured throughout the system with the security and privacy module (11). Using technologies such as Blockchain and FL (Federative Learning), security audits are performed throughout the entire system and innovative security protocols are implemented.

[0076] Finally, with the user interface and reporting module (12), users and administrators are provided with visualizations about system status, decisions made and performance. Thanks to the user interface and reporting module (12), users can provide feedback and take part in the continuous improvement of the system.

[0077] In the working method of the inventive system, in the data collection and preprocessing stages performed in the process steps of collecting raw data from physical networks (14) with at least one data collection module (1) (1001) and cleaning and preprocessing collected data with at least one data cleaning and pre-processing module (2) (1002), the flow diagram with additional security operations such as the use of blockchain and FL (Federated Learning) to ensure data integrity and security comprises the process steps of:

[0078] I. Securely collecting and transmitting data with at least one data collection module (1) (2001),

[0079] IL Ensuring data integrity using blockchain technologies and Federative Learning (FL) artificial intelligence model with at least one security and privacy module (11) (2002),

[0080] III. Using the pre-processed data to create virtual models with at least one simulation and modeling module (3) (2003),

[0081] IV. Analyzing data using artificial intelligence models (9) with at least one analysis module (4) (2004).

[0082] Said diagram illustrates the steps that must be taken to maintain the integrity of the data collected.

[0083] In the working method of the inventive system, a flow diagram specifying and detailing the process that enables the system to interact with the user in the step of presenting reports and visualizations to users with at least one user interface and reporting module (12) (1010) comprises the process steps of:I. Collecting data on system performance and status with at least one data collection module (1) (3001),

[0084] II. Creating reports and visualizations with at least one user interface and reporting module (12) (3002),

[0085] III. Presenting reports to users and administrators with at least one user interface and reporting module (12) (3003),

[0086] IV. Gathering and evaluating user feedback with at least one user interface and reporting module (12) (3004),

[0087] V. Improving the system based on feedback with optimization algorithms (15) (3005).

[0088] Said diagram shows a flow of user-oriented development of the system, such as performance monitoring, reporting, receiving feedback, and improving accordingly.

[0089] In the working method of the inventive system, the flow diagram, in which the step of making decisions about network optimization according to the results of the analysis by using optimization algorithms (15) with at least one decision-making module (7) (1007) is extended for efficiency, comprises the process steps of:

[0090] I. Selecting and implementing optimization algorithms (4001),

[0091] II. Optimizing resource allocation and energy efficiency (4002),

[0092] III. Determining performance improvement strategies (4003),

[0093] IV. Evaluating optimization results and adjustment if necessary (4004).

[0094] Said diagram shows the steps aimed to increase the energy efficiency of the network and improve resource utilization.

[0095] All modules in the system and method of the invention perform the operations mentioned in the invention over the processor included in the computer through a software.

[0096] Industrial Applicability of the Invention

[0097] The inventive system subject and the working method of this system will be used in various network types such as mobile networks (4G / 5G / 6G and beyond), wirelessnetworks, optical networks, satellite and aviation networks, vehicle networks (Vehicular Networks) and Industrial Internet of Things (HoT) networks, and are industrially applicable.

[0098] The invention is not limited to the example embodiments above, and the person skilled in the art can readily present other different embodiments of the invention. These should be considered within the protection scope of the invention claimed by the claims.

Claims

CLAIMS1. A computer-aided system (100) with at least one processor, aiming to increase network performance and security by performing artificial intelligence-based traffic analysis within Digital Twin Networks (DTN), characterized in that it comprises:at least one data collection module (1) that collects raw data, at least one data cleaning and pre-processing module (2) that cleans and performs preprocessing steps of the collected data, at least one simulation and modeling module (3) that creates models and simulations using the pre-processed data, at least one analysis module (4) that performs analysis with various artificial intelligence techniques, at least one prediction module (5) that predicts future network trends and potential problems, at least one detection module (6) that detects anomalies and potential security threats, at least one decision-making module (7) that makes decisions for network optimization based on the analysis results, at least one response module (8) that applies the decision made to the physical network (14), at least one artificial intelligence model (9), at least one communication interface (10) that ensures data and commands flow between modules, at least one security and privacy module (11) that ensures data security and privacy, at least one user interface and reporting module (12) that provides users and administrators with reports and visualizations on the status of the system, its performance, and decisions made, at least one virtual network (13) which comprises virtual models of physical networks (14), at least one physical network (14) which is the source of the data and where the optimizations are applied, and optimization algorithms (15) which comprise algorithms for performance improvement and are used in the decision-making module (7).

2. The working method of a computer-aided system (100) with at least one processor, aiming to increase network performance and security by performing artificial intelligence-based traffic analysis within Digital Twin Networks (DTN), characterized in that it comprises the process steps of:- Collecting raw data from physical networks (14) with at least one data collection module (1) (1001),- Cleaning and pre-processing collected data with at least one data cleaning and pre-processing module (2) (1002),- Using the pre-processed data to create virtual models with at least one simulation and modeling module (3) (1003),- Analyzing data with various artificial intelligence models (9) through at least one analysis module (4) (1004),- Predicting future network trends and potential problems with at least one prediction module (5) (1005),- Detecting anomalies and security threats in the network with at least one detection module (6) (1006),- Making decisions about network optimization according to the results of the analysis by using optimization algorithms (15) with at least one decision-making module (7) (1007),- Implementing decisions made with at least one response module (8) to the physical network (14) via the communication interface (10) (1008), - Ensuring data security and privacy with at least one security and privacy module (11) (1009),- Presenting reports and visualizations to users with at least one user interface and reporting module (12) (1010).

3. The working method according to claim 2, characterized in that, in the data collection and pre-processing stages performed in the process steps of collecting raw data from physical networks (14) with at least one data collection module (1) (1001) and cleaning and pre-processing collected data with at least one data cleaning and pre-processing module (2) (1002), the following process steps are applied which include additional security operations such as the use of blockchain and FL (Federated Learning) to ensure data integrity and security,- Securely collecting and transmitting data with at least one data collection module (1) (2001),- Ensuring data integrity using blockchain technologies and Federative Learning (FL) artificial intelligence model with at least one security and privacy module (11) (2002),- Using the pre-processed data to create virtual models with at least one simulation and modeling module (3) (2003),Analyzing data using artificial intelligence models (9) with at least one analysis module (4) (2004).

4. The working method of claim 2, characterized in that, in the step of presenting reports and visualizations to users with at least one user interface and reporting module (12) (1010), the following process steps are applied which specify and detail the process that enables the system to interact with the user,- Collecting data on system performance and status with at least one data collection module (1) (3001),- Creating reports and visualizations with at least one user interface and reporting module (12) (3002),- Presenting reports to users and administrators with at least one user interface and reporting module (12) (3003),- Gathering and evaluating user feedback with at least one user interface and reporting module (12) (3004),- Improving the system based on feedback with optimization algorithms (15) (3005).

5. The working method according to claim 2, characterized in that the following process steps are applied, in which the step of making decisions about network optimization according to the results of the analysis by using optimization algorithms (15) with at least one decision-making module (7) (1007) is extended for efficiency,- Selecting and implementing optimization algorithms (4001),- Optimizing resource allocation and energy efficiency (4002),- Determining performance improvement strategies (4003),- Evaluating optimization results and adjustment if necessary (4004).

Images