An AI-based intelligent and flexible routing activity rule system

Abstract

The present invention relates to the field of network technology, and in particular to a system based on AI intelligent flexible routing activity rules. It includes a system initialization and parameter configuration subsystem, an intelligent decision generation and adaptation subsystem, a rule execution and routing table adjustment subsystem, and a feedback and continuous optimization subsystem. From data acquisition, intelligent analysis, rule generation to execution feedback, the entire system of the present invention is highly automated. This greatly reduces the need for manual intervention and improves the operating efficiency and stability of the system. The AI ​​algorithm model built into the system can simulate the decision-making process of human experts and perform intelligent routing selection and rule adjustment. This intelligent decision-making capability enables the system to quickly make the best decision when facing a complex network environment.

Description

Technical Field

[0001] The present invention relates to the field of network technology, and in particular to an AI-based intelligent and flexible routing activity rule system. Background Art

[0002] Traditional routing systems mainly rely on preset routing rules to forward data packets, which lacks flexibility and intelligence. With the continuous development of network technology, network traffic and user needs are becoming increasingly complex and changeable, and traditional routing systems can no longer meet the needs of efficient and intelligent routing.

[0003] In modern communication networks, routing systems are a key component to ensure that data packets can be efficiently and accurately transmitted from source nodes to destination nodes. Traditional routing systems mainly rely on static preset routing rules and strategies, which are usually based on the administrator's experience and prior knowledge. In these systems, routing algorithms such as Dijkstra algorithm and Bellman-Ford algorithm are widely used to determine the transmission path of data packets by calculating the shortest path, minimum cost, etc.

[0004] However, with the rapid development of the Internet and other large-scale networks, network traffic patterns have become increasingly complex and changeable. The diversity of user needs, the diversity of application types (such as video streaming, online games, IoT devices, etc.), and the dynamic changes in the network environment (such as link status updates, node failures, etc.) make it difficult for traditional routing systems to meet the needs of efficient and intelligent routing in real time.

[0005] In the face of these challenges, some existing solutions have begun to try to achieve more flexible routing selection through dynamic routing protocols and policies. For example:

[0006] Dynamic routing protocols: Protocols such as OSPF (Open Shortest Path First) and BGP (Border Gateway Protocol) enable routers to dynamically adjust routing tables by periodically updating and transmitting network status information. However, these protocols are still based on preset strategies and have limited flexibility and intelligence.

[0007] Traffic Engineering: This method improves the utilization efficiency of network resources by analyzing and optimizing the distribution of network traffic. For example, MPLS (Multiprotocol Label Switching) traffic engineering technology can optimize bandwidth resources in large-scale networks, but it still operates within a pre-defined policy framework and lacks autonomous learning and intelligent decision-making capabilities.

[0008] Use SDN (Software Defined Network): SDN provides a high level of programmability and flexibility by separating the control plane from the data plane, allowing network administrators to dynamically adjust traffic paths based on demand. However, SDN usually requires high deployment and maintenance costs and relies on manual configuration by administrators, with a low degree of intelligence.

[0009] Although the above methods have improved the flexibility and efficiency of network routing to a certain extent, they still cannot fully cope with the complex and changing needs of modern networks. The lack of intelligent real-time decision-making capabilities is still the main bottleneck of the current routing system. Therefore, there is an urgent need for a new system based on AI intelligent and flexible adjustment of routing activity rules, which can achieve efficient and intelligent packet routing selection and dynamic adjustment by using artificial intelligence technology to meet the increasingly complex network needs. Summary of the invention

[0010] In view of the problems existing in the background technology, an AI-based intelligent and flexible routing activity rule system is proposed. Through key technologies such as real-time data collection, intelligent analysis, dynamic routing rule generation and adjustment, efficient rule execution, and closed-loop feedback learning, a highly automated and intelligent routing activity rule system is constructed, which provides new ideas and methods for the development and application of network technology.

[0011] The present invention proposes an AI-based intelligent flexible routing activity rule system, comprising a system initialization and parameter configuration subsystem, an intelligent decision generation and adaptation subsystem, a rule execution and routing table adjustment subsystem, and a feedback and continuous optimization subsystem; the system initialization and parameter configuration subsystem collects real-time network status data from various network devices through a data acquisition module and performs preliminary preprocessing; the intelligent decision generation and adaptation subsystem performs in-depth analysis on the collected network status data through a built-in AI algorithm model in the intelligent decision module; based on the real-time analyzed network status, the intelligent decision module automatically and instantly generates the optimal routing activity rules to adapt to the rapid changes in network conditions; the routing rule generation module dynamically adjusts the routing rules according to the changes in the network status; the rule execution module of the rule execution and routing table adjustment subsystem is integrated with the network device to convert the routing activity rules generated by the intelligent decision module into actual routing table items; the real-time control module controls the forwarding path of the data packet and optimizes resource allocation; the feedback and continuous optimization subsystem collects the actual effect data after the rule execution through the learning feedback module; the machine learning optimization module iteratively updates the AI ​​algorithm model.

[0012] Preferably, the system initialization and parameter configuration subsystem includes a network parameter configuration module, a data collection strategy module, a data storage module, a data collection module, and a system log and monitoring module. The network parameters configured by the network parameter configuration module include the IP address, subnet mask, gateway, and DNS server of the network device; the data collection strategy module is used to determine the frequency and range of data collection; the data storage module is used to store the storage space for network status data and historical data, using a database or a distributed file system; the data collection module starts to collect real-time network status data from each network device and performs preliminary preprocessing; the system log and monitoring module has system log and monitoring functions, records the system operation status and data collection details, and ensures that the entire system is correct during the initialization process.

[0013] Preferably, the intelligent decision generation and adaptation subsystem includes an intelligent decision module, a data input and preprocessing module, an AI algorithm selection module, a feature extraction and modeling module, a routing rule generation module, and a real-time update module. The intelligent decision module is the core part of the system, which performs in-depth analysis on the collected network status data through the AI ​​algorithm; the data input and preprocessing module will receive the real-time data transmitted by the data acquisition module, and perform data cleaning and formatting; the AI ​​algorithm selection module selects the most suitable AI algorithm for the current environment based on historical data and current network conditions; the feature extraction and modeling module extracts key features through feature engineering and establishes a prediction model; the routing rule generation module generates new optimal routing activity rules based on the prediction results and optimization goals; the real-time update module ensures that the update process is fully automated and carried out in real time, so that the generated rules can quickly respond to changes in the network environment and provide instant response.

[0014] Preferably, the rule execution and routing table adjustment subsystem includes a rule parsing and conversion module, a network device communication module, a routing table real-time adjustment module, and a configuration verification and rollback mechanism module. The rule parsing and conversion module parses the high-level rules generated by the intelligent decision-making module and converts them into specific network device configuration commands; the network device communication module communicates with each network device through API and network management protocol; the routing table real-time adjustment module adjusts the routing table of the network device according to the new configuration command and changes the forwarding path of the data packet; the configuration verification and rollback mechanism module verifies the effect of the new routing rules by real-time monitoring of the network status; if an abnormality occurs, the system has an automatic rollback mechanism to restore the route to the state before the adjustment.

[0015] Preferably, the feedback and continuous optimization subsystem includes an effect monitoring module, a data collection and analysis module, a machine learning optimization module, a model update and verification module, and a closed-loop design module. The effect monitoring module is used to continuously monitor the actual effect after the execution of the routing rules; the data collection and analysis module is used to collect comparison data before and after execution, analyze the impact of routing adjustments, and find out the difference between the actual effect and the expected goal; the machine learning optimization module uses the collected actual data and difference analysis results to retrain and optimize the machine learning model; the model update and verification module regularly updates the AI ​​model, and verifies the AI ​​model through simulation testing and actual deployment; the closed-loop design module forms a closed loop for the entire feedback learning process, and continuously improves and optimizes the system's intelligent decision-making ability and accuracy.

[0016] Compared with the prior art, the present invention has the following beneficial technical effects:

[0017] 1. High flexibility and intelligence:

[0018] Dynamic adaptability: The system can collect and analyze network status data in real time, and dynamically adjust routing activity rules according to changes in network traffic, fluctuations in node load, and differences in link quality. This high degree of flexibility enables the system to respond quickly to changes in the network environment and ensure that data packets are always transmitted through the optimal path.

[0019] Intelligent decision-making: The built-in AI algorithm model can simulate the decision-making process of human experts, conduct deep learning and intelligent analysis of network status, predict network trends and generate optimized routing rules. This intelligent decision-making capability not only improves the accuracy and efficiency of routing, but also reduces the need for human intervention.

[0020] 2. Improve network performance and efficiency:

[0021] Optimize resource allocation: By intelligently analyzing network status, the system can reasonably allocate network resources to avoid network congestion and overload. This helps to improve the speed and stability of data transmission and reduce packet loss and delay.

[0022] Reduce fault recovery time: When a network failure or anomaly is detected, the system can quickly adjust the routing rules to bypass the fault area and ensure the normal transmission of data packets. This rapid response capability reduces fault recovery time and improves network reliability and availability.

[0023] 3. Reduce operating costs and maintenance difficulty:

[0024] Automated management: The system realizes a fully automated process from data collection, intelligent analysis, rule generation to execution feedback, greatly reducing the need for manual intervention. This not only reduces the workload of operation and maintenance personnel, but also reduces the risk of human error.

[0025] Continuous optimization: The feedback learning module continuously optimizes the AI ​​algorithm model by collecting actual effect data after rule execution. This self-learning and iterative update capability enables the system to continuously adapt to new network environments, maintain its efficiency and accuracy, and thus reduce long-term operating costs.

[0026] 4. Enhance user experience:

[0027] Improve service quality: Through intelligent routing selection and dynamic adjustment rules, the system can ensure fast and stable transmission of user data packets. This helps to improve user experience and enhance user satisfaction and loyalty.

[0028] Support diverse needs: With the continuous development of network technology and the increasing diversification of user needs, the system can flexibly respond to various complex network scenarios and user needs, and provide more personalized and customized services. DETAILED DESCRIPTION

[0029] Embodiment 1

[0030] The present invention proposes an AI-based intelligent flexible routing activity rule system, which includes a system initialization and parameter configuration subsystem, an intelligent decision generation and adaptation subsystem, a rule execution and routing table adjustment subsystem, and a feedback and continuous optimization subsystem.

[0031] The system initialization and parameter configuration subsystem collects real-time network status data from various network devices through the data acquisition module and performs preliminary preprocessing, such as filtering abnormal data and filling missing data.

[0032] The intelligent decision generation and adaptation subsystem uses the built-in AI algorithm model in the intelligent decision module to conduct in-depth analysis of the collected network status data; based on the real-time analysis of the network status, the intelligent decision module automatically and instantly generates the optimal routing activity rules to adapt to the rapid changes in network conditions; the routing rule generation module dynamically adjusts the routing rules according to changes in the network status to ensure that the best routing path can be selected at any time, thereby improving the efficiency and reliability of data transmission.

[0033] The rule execution module of the rule execution and routing table adjustment subsystem is integrated with the network equipment to convert the routing activity rules generated by the intelligent decision-making module into actual routing table items to ensure efficient execution of the rules; the real-time control module controls the forwarding path of the data packet, and the system can effectively avoid network congestion, optimize resource allocation, and improve overall network performance.

[0034] The feedback and continuous optimization subsystem collects actual effect data after rule execution through the learning feedback module, including key indicators such as packet transmission time, packet loss rate, and delay, providing data support for the optimization of the AI ​​algorithm model; the machine learning optimization module iteratively updates the AI ​​algorithm model to continuously improve its prediction accuracy and decision-making accuracy.

[0035] The system initialization and parameter configuration subsystem includes a network parameter configuration module, a data collection strategy module, a data storage module, a data collection module, and a system log and monitoring module. The network parameters configured by the network parameter configuration module include basic network configurations such as the IP address, subnet mask, gateway, and DNS server of the network device; the data collection strategy module is used to determine the frequency and scope of data collection, such as collecting the traffic load, delay, and packet loss rate of each network device; the data storage module is used to store the storage space for network status data and historical data, using a database or distributed file system; the data collection module begins to collect real-time network status data from each network device and performs preliminary preprocessing, such as filtering abnormal data and filling in missing data; the system log and monitoring module has system log and monitoring functions, recording the system operation status and data collection details to ensure that the entire system is correct during the initialization process.

[0036] The intelligent decision generation and adaptation subsystem includes an intelligent decision module, a data input and preprocessing module, an AI algorithm selection module, a feature extraction and modeling module, a routing rule generation module, and a real-time update module.

[0037] The intelligent decision-making module is the core part of the system. It uses AI algorithms to conduct in-depth analysis of the collected network status data. The data input and preprocessing module will receive real-time data from the data acquisition module, perform data cleaning and formatting, and ensure the quality of the input data. The AI ​​algorithm selection module selects the most suitable AI algorithm for the current environment (such as machine learning, deep learning, reinforcement learning, etc.) based on historical data and current network conditions. The feature extraction and modeling module extracts key features through feature engineering and establishes prediction models, such as network traffic models, congestion prediction models, path optimization models, etc. The routing rule generation module generates new optimal routing activity rules based on prediction results and optimization goals (such as minimizing latency, maximizing bandwidth utilization, etc.); the real-time update module ensures that the update process is fully automated and carried out in real time, so that the generated rules can quickly respond to changes in the network environment and provide instant response.

[0038] The rule execution and routing table adjustment subsystem includes a rule parsing and conversion module, a network device communication module, a routing table real-time adjustment module, and a configuration verification and rollback mechanism module. The rule parsing and conversion module parses the high-level rules generated by the intelligent decision-making module and converts them into specific network device configuration commands (such as OpenFlow rules, router ACL, etc.); the network device communication module communicates with each network device through API and network management protocols (such as SNMP, NETCONF, etc.) to ensure the accuracy of configuration commands; the routing table real-time adjustment module adjusts the routing tables of network devices such as routers and switches according to the new configuration commands, changes the forwarding path of data packets, and achieves the purpose of optimizing network traffic; the configuration verification and rollback mechanism module verifies the effect of the new routing rules by real-time monitoring of the network status; if an abnormality occurs, the system has an automatic rollback mechanism to restore the route to the state before the adjustment to ensure the stability of the network.

[0039] The feedback and continuous optimization subsystem includes effect monitoring module, data collection and analysis module, machine learning optimization module, model updating and verification module and closed-loop design module.

[0040] The effect monitoring module is used to continuously monitor the actual effect after the execution of routing rules, including key performance indicators such as network delay, packet loss rate, and bandwidth utilization; the data collection and analysis module is used to collect comparative data before and after execution, analyze the impact of routing adjustments, and find out the difference between the actual effect and the expected goal; the machine learning optimization module uses the collected actual data and difference analysis results to retrain and optimize the machine learning model to make the AI ​​algorithm more accurate; the model update and verification module regularly updates the AI ​​model and verifies it through simulation tests and actual deployment, so that the model can perform well in different network environments; the closed-loop design module forms a closed loop for the entire feedback learning process, continuously improves and optimizes the system's intelligent decision-making ability and accuracy, and ensures that the system becomes more and more intelligent during operation.

[0041] Embodiment 2

[0042] The present invention proposes a working method based on an AI intelligent flexible routing activity rule system, which adopts the AI ​​intelligent flexible routing activity rule system described in Example 1, and the steps are as follows:

[0043] S1. System initialization and parameter configuration

[0044] When the system starts, it first initializes and configures the parameters of each module:

[0045] S11. Configure network parameters: including basic network configurations such as the IP address, subnet mask, gateway, DNS server, etc. of network devices.

[0046] S12. Define data collection strategy: Determine the frequency and scope of data collection, such as collecting the traffic load, latency, packet loss rate, etc. of each network device.

[0047] S13. Initialize data storage: allocate storage space for storing network status data and historical data, using a database or distributed file system.

[0048] S14, start the data collection module: This module starts to collect real-time network status data from various network devices and performs preliminary preprocessing, such as filtering abnormal data and filling missing data.

[0049] S15, System log and monitoring: Enable the system log and monitoring functions to record the system operation status and data collection details to ensure that the entire system is correct during the initialization process.

[0050] S2. Intelligent decision generation and adaptation

[0051] The intelligent decision-making module is the core part of the system, which uses AI algorithms to conduct in-depth analysis of the collected network status data:

[0052] S21. Data input and preprocessing: Receive real-time data from the data acquisition module, perform data cleaning and formatting to ensure the quality of input data.

[0053] S22. AI algorithm selection: Based on historical data and current network conditions, select the AI ​​algorithm that best suits the current environment (such as machine learning, deep learning, reinforcement learning, etc.).

[0054] S23. Feature extraction and modeling: Extract key features through feature engineering and establish prediction models, such as network traffic models, congestion prediction models, path optimization models, etc.

[0055] S24, generate routing rules: generate new optimal routing activity rules based on the prediction results and optimization goals (such as minimizing delay, maximizing bandwidth utilization, etc.).

[0056] S25. Real-time update: Ensure that this process is fully automated and carried out in real time so that the generated rules can quickly respond to changes in the network environment and provide immediate response.

[0057] S3, rule execution and routing table adjustment

[0058] After the rules are generated, they need to be applied to the actual network devices immediately:

[0059] S31, rule parsing and conversion: Parse the high-level rules generated by the intelligent decision-making module and convert them into specific network device configuration commands (such as OpenFlow rules, router ACL, etc.).

[0060] S32. Network device communication: Communicate with each network device through API and network management protocols (such as SNMP, NETCONF, etc.) to ensure the accuracy of configuration commands.

[0061] S33. Real-time adjustment of routing tables: According to new configuration commands, the routing tables of network devices such as routers and switches are quickly adjusted to change the forwarding path of data packets to achieve the purpose of optimizing network traffic.

[0062] S34, Configuration verification and rollback mechanism: Verify the effect of new routing rules by real-time monitoring of network status. If an abnormality occurs, the system has an automatic rollback mechanism to restore the route to the state before adjustment to ensure network stability.

[0063] S4. Feedback and Continuous Optimization

[0064] In order to enable the system to have self-learning and self-optimization capabilities, the feedback learning module plays a key role:

[0065] S41. Effect monitoring: Continuously monitor the actual effect of routing rules after execution, including key performance indicators such as network latency, packet loss rate, and bandwidth utilization.

[0066] S42. Data collection and analysis: Collect comparative data before and after execution, analyze the impact of route adjustment, and find out the difference between actual effect and expected goal.

[0067] S43. Machine learning optimization: Use the collected actual data and difference analysis results to retrain and optimize the machine learning model to make the AI ​​algorithm more accurate.

[0068] S44. Model update and verification: Update AI models regularly and verify them through simulation tests and actual deployment so that the models can perform well in different network environments.

[0069] S45. Closed-loop design: The entire feedback learning process forms a closed loop, continuously improving and optimizing the system's intelligent decision-making capabilities and accuracy, ensuring that the system becomes more and more intelligent during operation.

[0070] The present invention realizes a high degree of automation in the entire system from data collection, intelligent analysis, rule generation to execution feedback. This greatly reduces the need for manual intervention and improves the operating efficiency and stability of the system. The AI ​​algorithm model built into the system can simulate the decision-making process of human experts and perform intelligent routing selection and rule adjustment. This intelligent decision-making capability enables the system to quickly make the best decision when facing a complex network environment.

[0071] The above describes the embodiments of the present invention in detail, but the present invention is not limited thereto. Various changes can be made within the knowledge scope of those skilled in the art without departing from the spirit of the present invention.

Claims

1. A system based on AI intelligent and flexible routing activity rules, characterized in that: It includes system initialization and parameter configuration subsystem, intelligent decision generation and adaptation subsystem, rule execution and routing table adjustment subsystem, and feedback and continuous optimization subsystem; The system initialization and parameter configuration subsystem collects real-time network status data from various network devices through the data acquisition module and performs preliminary preprocessing; The intelligent decision generation and adaptation subsystem uses the built-in AI algorithm model in the intelligent decision module to conduct in-depth analysis of the collected network status data; based on the real-time analysis of the network status, the intelligent decision module automatically and instantly generates the optimal routing activity rules to adapt to the rapid changes in network conditions; the routing rule generation module dynamically adjusts the routing rules according to the changes in the network status; The rule execution module of the rule execution and routing table adjustment subsystem is integrated with the network equipment to convert the routing activity rules generated by the intelligent decision module into actual routing table entries; the real-time control module controls the forwarding path of the data packet and optimizes resource allocation; The feedback and continuous optimization subsystem collects actual effect data after rule execution through the learning feedback module; the machine learning optimization module iteratively updates the AI ​​algorithm model.

2. The AI ​​intelligent flexible routing activity rule system according to claim 1 is characterized in that: The system initialization and parameter configuration subsystem includes: A network parameter configuration module, the network parameters configured by the network parameter configuration module include the IP address, subnet mask, gateway and DNS server of the network device; Data collection strategy module, which is used to determine the frequency and scope of data collection; Data storage module: the data storage module is used to store network status data and historical data, using a database or a distributed file system; Data collection module, which starts to collect real-time network status data from various network devices and performs preliminary preprocessing; And the system log and monitoring module, which has system log and monitoring functions, records the system operation status and data collection details, and ensures that the entire system is correct during the initialization process.

3. The AI ​​intelligent flexible routing activity rule system according to claim 1 is characterized in that: The intelligent decision generation and adaptation subsystem includes: Intelligent decision-making module: The intelligent decision-making module is the core part of the system. It uses AI algorithms to conduct in-depth analysis of the collected network status data. Data input and preprocessing module: The data input and preprocessing module will receive the real-time data from the data acquisition module and perform data cleaning and formatting; AI algorithm selection module: The AI ​​algorithm selection module selects the most suitable AI algorithm for the current environment based on historical data and current network conditions; Feature extraction and modeling module, which extracts key features through feature engineering and builds a prediction model; Routing rule generation module, which generates new optimal routing activity rules based on prediction results and optimization goals; And the real-time update module ensures that the update process is fully automated and carried out in real time, so that the generated rules can quickly adapt to changes in the network environment and provide immediate response.

4. The AI ​​intelligent flexible routing activity rule system according to claim 1 is characterized in that: The rule execution and routing table adjustment subsystem includes: The rule parsing and conversion module parses the high-level rules generated by the intelligent decision-making module and converts them into specific network device configuration commands; Network device communication module, which communicates with various network devices through API and network management protocol; A real-time routing table adjustment module, which adjusts the routing table of network devices according to new configuration commands and changes the forwarding path of data packets; As well as a configuration verification and rollback mechanism module, this module verifies the effectiveness of new routing rules by monitoring the network status in real time; if an abnormality occurs, the system has an automatic rollback mechanism to restore the route to the state before the adjustment.

5. The AI ​​intelligent flexible routing activity rule system according to claim 1 is characterized in that: The feedback and continuous optimization subsystem includes: Effect monitoring module, which is used to continuously monitor the actual effect of routing rules after execution; Data collection and analysis module, which is used to collect comparison data before and after execution, analyze the impact of route adjustment, and find out the difference between actual effect and expected goal; The machine learning optimization module uses the collected actual data and the difference analysis results to retrain and optimize the machine learning model; Model update and verification module, which regularly updates AI models and verifies them through simulation tests and actual deployment; And a closed-loop design module, which closes the entire feedback learning process to continuously improve and optimize the system's intelligent decision-making capabilities and accuracy.