Method for multi-path cooperative network acceleration
By real-time monitoring and integration of multi-path network traffic data, combined with machine learning and optimization algorithms, the problem of data acquisition device calibration status detection was solved, and the accuracy and stability of the multi-path collaborative network acceleration method were improved, thereby enhancing network performance and resource utilization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- RAYTHEON (WUHAN) NETWORK TECH CO LTD
- Filing Date
- 2026-04-21
- Publication Date
- 2026-06-23
AI Technical Summary
In existing technologies, multi-path cooperative network acceleration methods cannot detect the calibration status of data acquisition devices in real time during data acquisition, resulting in data deviation and delay, which affects the accuracy of traffic feature analysis and path performance evaluation, and consequently affects the accuracy of strategy selection.
By collecting multi-path network traffic data and path status data, real-time monitoring and integration are performed. Machine learning and weighted scoring models are used for traffic feature extraction and path performance evaluation. Multi-objective optimization algorithms and linear programming models are combined for resource allocation to accelerate data transmission. Furthermore, through real-time path switching and load balancing, artificial intelligence-assisted decision-making is integrated to generate intelligent decision data.
It enables real-time calibration of multi-path network traffic data, improves the accuracy of traffic characteristics and path performance evaluation, ensures the real-time nature of resource allocation and the stability of network acceleration, and enhances the overall performance and resource utilization of multi-path cooperative networks.
Smart Images
Figure CN122268809A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of network communication technology, specifically a method for accelerating multi-path cooperative networks. Background Technology
[0002] A network is a physical link that connects isolated workstations and hosts to form a data link, thereby achieving the purpose of resource sharing and communication. Communication is the exchange and transmission of information between people through some medium. Network communication connects isolated devices through a network and realizes communication between people, between people and computers, and between computers through information exchange.
[0003] Currently, because multi-path cooperative network acceleration methods involve various dynamic network parameters during implementation, it is impossible to detect in real time whether the calibration status of the data acquisition equipment is abnormal when collecting and integrating multi-path network traffic data and path status data. When the collected data has deviations and delays, it will cause large errors in multi-path traffic feature analysis and path performance evaluation, and cannot guarantee the accuracy of subsequent strategy selection.
[0004] Therefore, a multi-path cooperative network acceleration method is proposed to solve the above problems. Summary of the Invention
[0005] To address the shortcomings of existing technologies, this invention provides a multi-path cooperative network acceleration method, which solves the problems mentioned in the background section regarding the inability to detect abnormal calibration status of data acquisition devices in real time and the inability to guarantee the accuracy of subsequent strategy selection.
[0006] To achieve the above objectives, the present invention provides the following technical solution: a multi-path cooperative network acceleration method, the method comprising the following steps: S1. Collect multi-path network traffic data and path status data; S2. Perform multi-path network traffic analysis and path performance evaluation processing, and generate multi-path traffic feature data and path performance evaluation data based on the multi-path network traffic data and path status data. S3. Based on the multi-path traffic characteristic data and path performance evaluation data, perform multi-path collaborative strategy selection processing to generate multi-path collaborative strategy data; S4. Perform path resource allocation optimization processing based on the multi-path collaboration strategy data to generate optimized path resource allocation data; S5. Based on the optimized path resource allocation data, perform multi-path data acceleration transmission processing to generate multi-path accelerated transmission data; S6. Perform multi-path collaboration effect monitoring and dynamic adjustment processing, and generate multi-path collaboration monitoring data and dynamic adjustment instructions based on the multi-path accelerated transmission data. S7. Based on the multi-path collaborative monitoring data and dynamic adjustment instructions, perform real-time path switching and load balancing to generate final network acceleration data. S8. Based on the final network acceleration data, perform long-term optimization processing of the multi-path cooperative network, including historical data analysis and policy library update, to generate long-term optimization data; S9. Integrate artificial intelligence-assisted decision processing, and generate intelligent decision data based on the long-term optimized data by using deep learning prediction and reinforcement learning adjustment.
[0007] Preferably, the collection of multi-path network traffic data and path status data in S1 specifically includes the following steps: S11. Collect multi-path network traffic data through distributed network sensors, including packet size, transmission rate and latency parameters; S12. Collect path status data through path status monitoring equipment, including path available bandwidth, packet loss rate and jitter parameters; S13. Integrate the multi-path network traffic data and path status data into the network management platform, and generate a multi-path network traffic dataset and a path status dataset. The data acquisition quality is calculated using the following formula: ; in, The data collection quality score is used to assess the completeness and reliability of the collected data. The number of valid data points. This represents the total number of data points.
[0008] Preferably, the multi-path network traffic analysis and path performance evaluation processing in step S2, and the generation of multi-path traffic feature data and path performance evaluation data based on the multi-path network traffic data and path status data, specifically includes the following steps: S21. Based on the multi-path network traffic dataset, perform traffic feature extraction processing, use machine learning algorithms to identify traffic patterns, and generate multi-path traffic feature data; S22. Based on the path status dataset, perform path performance evaluation processing, calculate the performance score of each path using a weighted scoring model, and generate path performance evaluation data. Performance scores are calculated using the following formula: ; in, To score path performance and quantify the quality of a path, , , For the weighting coefficients, satisfying , The normalized path available bandwidth. The normalized path delay parameter, This represents the normalized path packet loss rate. S23. The multi-path traffic characteristic data and path performance evaluation data are associated and stored to construct a multi-path performance analysis database.
[0009] Preferably, the multi-path collaborative strategy selection process in S3, based on the multi-path traffic characteristic data and path performance evaluation data, to generate multi-path collaborative strategy data specifically includes the following steps: S31. Obtain the multi-path traffic characteristic data and path performance evaluation data; S32. A multi-objective optimization algorithm is used to select multi-path collaborative strategies, and multi-path collaborative strategy data is generated based on traffic priority and path stability. S33. When the path performance evaluation data is lower than the threshold, the adaptive strategy adjustment mechanism is triggered to regenerate the multi-path collaborative strategy data.
[0010] Preferably, step S4, which optimizes path resource allocation based on the multi-path collaboration strategy data to generate optimized path resource allocation data, specifically includes the following steps: S41. Based on the multi-path collaborative strategy data, perform path resource allocation optimization processing, and use a linear programming model to allocate bandwidth and computing resources. S42. Generate optimized path resource allocation data, including resource quotas and scheduling schedules for each path; S43. Ensure the real-time availability of path resources through the resource reservation protocol, and update and optimize path resource allocation data.
[0011] Preferably, in step S5, multi-path data acceleration transmission processing is performed based on the optimized path resource allocation data to generate multi-path accelerated transmission data, specifically including the following steps: S51. Implement multi-path data acceleration transmission processing based on the optimized path resource allocation data, using data fragmentation and parallel transmission techniques. S52. Perform network packet scheduling processing, and generate multi-path accelerated transmission data based on priority queues and congestion control algorithms; S53. Enhances data transmission security through encryption and compression modules, and accelerates the integrity of transmitted data through real-time verification.
[0012] Preferably, in step S6, multi-path collaboration effect monitoring and dynamic adjustment processing are performed. Based on the multi-path accelerated transmission data, generating multi-path collaboration monitoring data and dynamic adjustment instructions specifically includes the following steps: S61. Perform multi-path collaboration effect monitoring and processing, and generate multi-path collaboration monitoring data through real-time data stream analysis; S62. Generate dynamic adjustment instructions based on the deviation between monitoring data and preset benchmarks, including path weight adjustment and traffic rerouting instructions; S63. When a path failure is detected, the fault recovery mechanism is automatically triggered to update the multi-path collaborative monitoring data and dynamic adjustment instructions.
[0013] Preferably, in step S7, the real-time path switching and load balancing processing based on the multi-path collaborative monitoring data and dynamic adjustment instructions to generate the final network acceleration data specifically includes the following steps: S71. Based on the multi-path collaborative monitoring data and dynamic adjustment instructions, perform real-time path switching processing and use a fast switching algorithm to minimize latency; S72. Implement load balancing to dynamically distribute network traffic to the optimal path and generate final network acceleration data. S73. The acceleration effect analysis is output through the log recording and reporting module, which is used by network administrators for subsequent optimization.
[0014] Preferably, in step S8, long-term optimization processing of the multi-path cooperative network is performed based on the final network acceleration data, including historical data analysis and policy base updates. The specific steps for generating long-term optimization data are as follows: S81. Regularly collect historical acceleration data to perform trend analysis and pattern recognition; S82. Update the multi-path collaboration strategy library based on the analysis results; S83. Verify the optimization effect through simulation testing to ensure the robustness of the method under different network environments.
[0015] Preferably, step S9 integrates artificial intelligence-assisted decision processing, using deep learning prediction and reinforcement learning adjustment based on the long-term optimization data to generate intelligent decision data, specifically including the following steps: S91. Use deep learning models to predict network traffic changes and generate multi-path collaborative strategies in advance. S92. Adaptive acceleration is achieved by dynamically adjusting path selection parameters through reinforcement learning algorithms. S93. Combine AI decision-making results with human intervention.
[0016] Compared with existing technologies, this invention provides a multi-path cooperative network acceleration method, which has the following beneficial effects: 1. In this invention, when collecting and integrating multi-path network traffic data and path status data, the data collection process is monitored in real time, and the collected multi-path network traffic data and path status data are integrated into the network management platform for consistency verification. This enables the real-time identification of deviations in the data collection process, ensuring the accuracy of the generated multi-path traffic characteristic data and path performance evaluation data. This provides a reliable basis for the selection of multi-path collaborative strategies and improves the data foundation quality for multi-path collaborative network acceleration.
[0017] 2. In this invention, when performing multi-path collaboration strategy selection and path resource allocation optimization, the multi-path collaboration strategy data is compared with the current network status in real time, and the path resource allocation data is dynamically adjusted and optimized based on the multi-path collaboration monitoring data. This ensures that the resource allocation matches the real-time network conditions, avoids the failure of path resource allocation, and thus improves the efficiency of multi-path data acceleration transmission and resource utilization.
[0018] 3. In this invention, when performing multi-path data acceleration transmission processing, by implementing real-time path switching and load balancing mechanisms, the transmission path of multi-path accelerated data transmission can be dynamically optimized, reducing data packet transmission anomalies, thereby enhancing the stability and continuity of the final network accelerated data and improving the overall performance of multi-path collaborative network acceleration. Attached Figure Description
[0019] Figure 1 This is a flowchart illustrating the steps of the multi-path cooperative network acceleration method of the present invention. Detailed Implementation
[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0021] Please see Figure 1 The specific implementation of the multi-path cooperative network acceleration method is as follows, which includes the following steps: S1. Collect multi-path network traffic data and path status data; S2. Perform multi-path network traffic analysis and path performance evaluation processing, and generate multi-path traffic characteristic data and path performance evaluation data based on multi-path network traffic data and path status data. S3. Based on multi-path traffic characteristic data and path performance evaluation data, perform multi-path collaborative strategy selection processing to generate multi-path collaborative strategy data; S4. Optimize path resource allocation based on multi-path collaboration strategy data to generate optimized path resource allocation data; S5. Based on the optimized path resource allocation data, implement multi-path data acceleration transmission processing to generate multi-path accelerated transmission data; S6. Perform multi-path collaboration effect monitoring and dynamic adjustment processing, and generate multi-path collaboration monitoring data and dynamic adjustment instructions based on multi-path accelerated transmission data. S7. Based on multi-path collaborative monitoring data and dynamic adjustment instructions, perform real-time path switching and load balancing to generate final network acceleration data; S8. Perform long-term optimization processing of multi-path cooperative networks based on the final network acceleration data, including historical data analysis and policy library updates, to generate long-term optimization data; S9 integrates artificial intelligence-assisted decision processing, using deep learning prediction and reinforcement learning adjustment based on long-term optimized data to generate intelligent decision data.
[0022] The specific steps for collecting multi-path network traffic data and path status data in S1 are as follows: S11. Collect multi-path network traffic data through distributed network sensors, including packet size, transmission rate and latency parameters; S12. Collect path status data through path status monitoring equipment, including path available bandwidth, packet loss rate and jitter parameters; S13. Integrate multi-path network traffic data and path status data into the network management platform, and generate multi-path network traffic dataset and path status dataset; The data acquisition quality is calculated using the following formula: ; in, The data collection quality score is used to assess the completeness and reliability of the collected data. The number of valid data points. This represents the total number of data points.
[0023] In S2, multi-path network traffic analysis and path performance evaluation are performed. Based on multi-path network traffic data and path status data, multi-path traffic characteristic data and path performance evaluation data are generated. The specific steps include: S21. Based on the multi-path network traffic dataset, traffic feature extraction processing is performed, machine learning algorithms are used to identify traffic patterns, and multi-path traffic feature data is generated. S22. Based on the path status dataset, perform path performance evaluation processing, use a weighted scoring model to calculate the performance score of each path, and generate path performance evaluation data. Among them, the machine learning algorithm adopts the decision tree algorithm, which learns the pattern by training historical traffic data. Feature extraction includes calculating the mean and variance of data packet size and detecting peak transmission rate. The training data comes from public network datasets, and the hyperparameters are optimized through grid search. Performance scores are calculated using the following formula: ; in, To score path performance and quantify the quality of a path, , , For the weighting coefficients, satisfying , The normalized path available bandwidth. The normalized path delay parameter, This represents the normalized path packet loss rate. S23. Link and store multi-path traffic characteristic data and path performance evaluation data to construct a multi-path performance analysis database.
[0024] In S3, the process of selecting multi-path collaborative strategies based on multi-path traffic characteristic data and path performance evaluation data to generate multi-path collaborative strategy data includes the following steps: S31. Obtain multi-path traffic characteristic data and path performance evaluation data; S32. A multi-objective optimization algorithm is used to select multi-path collaborative strategies, and multi-path collaborative strategy data is generated based on traffic priority and path stability. S33. When the path performance evaluation data is lower than the threshold, the adaptive strategy adjustment mechanism is triggered to regenerate the multi-path collaborative strategy data.
[0025] In S4, the path resource allocation optimization process is performed based on the multi-path coordination strategy data. The specific steps to generate optimized path resource allocation data include: S41. Optimize path resource allocation based on multi-path collaborative strategy data, and use a linear programming model to allocate bandwidth and computing resources. The specific operations include: establishing resource allocation constraints, using available path bandwidth and latency as variables, and solving for the optimal resource allocation scheme; S42. Generate optimized path resource allocation data, including resource quotas and scheduling schedules for each path; Resource allocation efficiency is calculated using the following formula: ; in, To improve resource allocation efficiency and evaluate resource utilization effectiveness, For the number of paths, This represents the actual resource usage of the i-th path. Let be the weight coefficient of the i-th path. Total resource capacity; The specific operations include: collecting resource usage data for each path in real time, substituting it into the formula to calculate the efficiency value, and comparing it with the threshold to adjust the allocation strategy; S43. Ensure the real-time availability of path resources through the resource reservation protocol, and update and optimize path resource allocation data.
[0026] In S5, multi-path data acceleration transmission processing is implemented based on optimized path resource allocation data. The specific steps for generating multi-path accelerated transmission data include: S51. Implement multi-path data acceleration transmission processing based on optimized path resource allocation data, using data fragmentation and parallel transmission techniques; The specific operations include: fragmenting data packets according to priority and transmitting them simultaneously via multiple paths; S52. Perform network packet scheduling processing, and generate multi-path accelerated transmission data based on priority queues and congestion control algorithms; Transmission efficiency is calculated using the formula: ; in, To improve transmission efficiency, data transmission rate and success rate are measured. The amount of data successfully transmitted, Total data volume This represents the average transmission delay. Specific operations include: monitoring real-time data transmission, substituting data into formulas to calculate efficiency, and dynamically adjusting the fragment size and path selection to optimize efficiency; S53. Enhances data transmission security through encryption and compression modules, and accelerates the integrity of transmitted data through real-time verification.
[0027] In S6, multi-path collaboration effect monitoring and dynamic adjustment processing are performed. Based on multi-path accelerated data transmission, multi-path collaboration monitoring data and dynamic adjustment instructions are generated, specifically including the following steps: S61. Perform multi-path collaboration effect monitoring and processing, and generate multi-path collaboration monitoring data through real-time data stream analysis; Specific operations include: setting the monitoring cycle, collecting and storing path performance indicators; S62. Generate dynamic adjustment instructions based on the deviation between monitoring data and preset benchmarks, including path weight adjustment and traffic rerouting instructions; Monitoring accuracy is calculated using the following formula: ; in, To improve monitoring accuracy and ensure the accuracy of monitoring data assessment, These are actual monitored values. This is the expected monitoring value. This represents the maximum possible range of the monitored values. The specific operations include: periodically calibrating the monitoring equipment, substituting the values into the formula to calculate the accuracy value, and triggering the equipment to recalibrate when the accuracy is lower than the threshold. S63. When a path failure is detected, the fault recovery mechanism is automatically triggered to update the multi-path collaborative monitoring data and dynamic adjustment instructions.
[0028] S7 performs real-time path switching and load balancing based on multi-path collaborative monitoring data and dynamic adjustment commands to generate final network acceleration data. The specific steps include: S71. Real-time path switching is performed based on multi-path collaborative monitoring data and dynamic adjustment instructions, and a fast switching algorithm is used to minimize latency. Specific operations include: identifying bottleneck paths based on monitoring data and switching to backup paths within milliseconds; S72. Implement load balancing to dynamically distribute network traffic to the optimal path and generate final network acceleration data. Load balancing is calculated using the formula: ; in, To achieve load balancing and quantify the uniformity of traffic distribution, The standard deviation of the load for each path reflects the degree of load fluctuation. The standard deviation is calculated based on real-time sampling of the path load, with a sampling frequency of once per second. The specific operations include: calculating the load data of each path in real time, substituting it into the formula to calculate the balance, and adjusting the traffic weight to make the balance close to the ideal value. S73. The acceleration effect analysis is output through the log recording and reporting module, which is used by network administrators for subsequent optimization.
[0029] In S8, long-term optimization of multi-path cooperative networks is performed based on the final network acceleration data, including historical data analysis and policy base updates. The specific steps to generate long-term optimization data are as follows: S81. Regularly collect historical acceleration data to perform trend analysis and pattern recognition; Specific operations include: using time series analysis algorithms to identify patterns in performance fluctuations; S82. Update the multi-path collaborative strategy library based on the analysis results to improve the accuracy of future accelerated processing; Optimized gain is calculated using the formula: ; in, To optimize gain and evaluate long-term optimization effects, The optimized performance metric values, These are the performance metrics before optimization. The performance metrics are comprehensive scores, calculated based on a weighted average of latency, bandwidth, and packet loss rate. The specific operations include: comparing the data before and after optimization, substituting it into the formula to calculate the gain, and adjusting the update frequency of the strategy library based on the gain value; S83. Verify the optimization effect through simulation testing to ensure the robustness of the method under different network environments.
[0030] S9 integrates AI-assisted decision processing, using deep learning prediction and reinforcement learning adjustments based on long-term optimized data to generate intelligent decision data, specifically including the following steps: S91. A deep learning model is used to predict network traffic changes and generate multi-path collaborative strategies in advance. The deep learning model uses a long short-term memory network. The input is a historical traffic sequence and the output is a future traffic prediction. The Adam optimizer is used for training, and the learning rate is 0.001. S92. The path selection parameters are dynamically adjusted through the reinforcement learning algorithm to achieve adaptive acceleration. The reinforcement learning algorithm adopts Q-learning, the state space includes path load and latency, and the reward function is based on the transmission success rate and latency reduction. S93. Combine AI decision-making results with human intervention to improve the intelligence level of network acceleration.
[0031] The steps of the multi-path cooperative network acceleration method are as follows: Step 1: Collect multi-path network traffic data and path status data: This method begins by collecting multi-path network traffic data and path status data in real time using distributed network sensors and path status monitoring devices. The multi-path network traffic data includes packet size, transmission rate, and latency parameters, while the path status data covers path available bandwidth, packet loss rate, and jitter parameters. After collection, the data is integrated into the network management platform to generate multi-path network traffic datasets and path status datasets, providing a foundation for subsequent analysis. This step ensures the comprehensiveness and real-time nature of the data sources and avoids the problem of data silos through platform integration.
[0032] Step 2: Perform multi-path network traffic analysis and path performance evaluation. Based on the collected dataset, the system performs multi-path network traffic analysis and path performance evaluation. First, it extracts traffic features through machine learning algorithms to generate multi-path traffic feature data and identifies traffic patterns, including bursts and periods. Second, it uses a weighted scoring model to calculate the performance score of each path, generating path performance evaluation data to quantify the quality of paths. The processing results are stored in a multi-path performance analysis database to achieve data traceability. The core principle of this step is to use intelligent analysis to enhance the value of data and provide a quantitative basis for strategy selection.
[0033] Step 3: Select and process multi-path collaborative strategies based on multi-path traffic characteristic data and path performance evaluation data. In this step, the system performs multi-path collaborative strategy selection processing based on multi-path traffic characteristic data and path performance evaluation data. Through a multi-objective optimization algorithm, it comprehensively considers traffic priority and path stability to generate multi-path collaborative strategy data. When the path performance evaluation data is lower than the threshold, an adaptive strategy adjustment mechanism is triggered to regenerate the strategy data. In principle, this step emphasizes dynamic optimization to ensure that the strategy matches the network state and avoid resource allocation failure.
[0034] Step 4: Optimize path resource allocation based on multi-path collaboration strategy data. Based on multi-path coordination strategy data, the system performs path resource allocation optimization, uses a linear programming model to allocate bandwidth and computing resources, generates optimized path resource allocation data, including resource quotas and scheduling schedules, calculates resource allocation efficiency using formulas, evaluates utilization effects in real time, and combines resource reservation protocols to ensure resource availability. In principle, this step achieves resource scheduling through mathematical modeling, thereby improving overall efficiency.
[0035] Step 5: Implement multi-path data acceleration transmission processing based on optimized path resource allocation data: Based on optimized path resource allocation data, the system implements multi-path data acceleration transmission processing. It employs data fragmentation and parallel transmission techniques, combined with priority queues and congestion control algorithms, to generate multi-path accelerated transmission data. Transmission efficiency is quantified by a formula, dynamically adjusting fragment size and path selection to ensure high speed and high success rate. The principle behind this step is leveraging multi-path parallelism to reduce transmission latency and data loss.
[0036] Step Six: Monitor and dynamically adjust the multi-path collaboration effect: The system continuously monitors and dynamically adjusts the multi-path collaboration effect. It generates multi-path collaboration monitoring data through real-time data stream analysis and generates dynamic adjustment instructions based on deviations, including path weight adjustment. The monitoring accuracy is evaluated by formula, and when the accuracy is insufficient, the equipment is triggered to recalibrate. In principle, this step realizes closed-loop control and ensures the stability and adaptability of the system through monitoring feedback.
[0037] Step 7: Perform real-time path switching and load balancing based on multi-path collaborative monitoring data and dynamic adjustment commands: By leveraging multi-path collaborative monitoring data and dynamic adjustment commands, the system performs real-time path switching and load balancing, minimizes latency using a fast switching algorithm, and optimizes traffic distribution through a load balancing formula to generate final network acceleration data. In principle, this step focuses on responding to real-time changes and improving network resilience and user experience.
[0038] Step 8: Perform long-term optimization of the multi-path cooperative network based on the final network acceleration data: The system performs long-term optimization based on the final network acceleration data. Through historical data trend analysis and pattern recognition, it updates the multi-path collaborative strategy library and generates long-term optimization data. The optimization gain is calculated by formula to guide the update frequency of the strategy library. In principle, this step embodies the machine learning cycle and enables the system to self-evolve.
[0039] Step 9: Integrate AI-assisted decision processing: Finally, the system integrates artificial intelligence-assisted decision processing, using deep learning to predict traffic changes and reinforcement learning to adjust parameters to generate intelligent decision data. In principle, this step improves the level of intelligence, enabling the method to adapt to complex network environments and enhance the sustainability of the acceleration effect.
[0040] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0041] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A multi-path cooperative network acceleration method, characterized in that, The method includes the following steps: S1. Collect multi-path network traffic data and path status data; S2. Perform multi-path network traffic analysis and path performance evaluation processing, and generate multi-path traffic feature data and path performance evaluation data based on the multi-path network traffic data and path status data. S3. Based on the multi-path traffic characteristic data and path performance evaluation data, perform multi-path collaborative strategy selection processing to generate multi-path collaborative strategy data; S4. Perform path resource allocation optimization processing based on the multi-path collaboration strategy data to generate optimized path resource allocation data; S5. Based on the optimized path resource allocation data, perform multi-path data acceleration transmission processing to generate multi-path accelerated transmission data; S6. Perform multi-path collaboration effect monitoring and dynamic adjustment processing, and generate multi-path collaboration monitoring data and dynamic adjustment instructions based on the multi-path accelerated transmission data. S7. Based on the multi-path collaborative monitoring data and dynamic adjustment instructions, perform real-time path switching and load balancing to generate final network acceleration data. S8. Based on the final network acceleration data, perform long-term optimization processing of the multi-path cooperative network, including historical data analysis and policy library update, to generate long-term optimization data; S9. Integrate artificial intelligence-assisted decision processing, and generate intelligent decision data based on the long-term optimized data by using deep learning prediction and reinforcement learning adjustment.
2. The multi-path cooperative network acceleration method according to claim 1, characterized in that, The collection of multi-path network traffic data and path status data in S1 specifically includes the following steps: S11. Collect multi-path network traffic data through distributed network sensors, including packet size, transmission rate and latency parameters; S12. Collect path status data through path status monitoring equipment, including path available bandwidth, packet loss rate and jitter parameters; S13. Integrate the multi-path network traffic data and path status data into the network management platform, and generate a multi-path network traffic dataset and a path status dataset. The data acquisition quality is calculated using the following formula: ； in, The data collection quality score is used to assess the completeness and reliability of the collected data. The number of valid data points. This represents the total number of data points.
3. The multi-path cooperative network acceleration method according to claim 1, characterized in that, The S2 step involves multi-path network traffic analysis and path performance evaluation. Based on the multi-path network traffic data and path status data, the generation of multi-path traffic characteristic data and path performance evaluation data specifically includes the following steps: S21. Based on the multi-path network traffic dataset, perform traffic feature extraction processing, use machine learning algorithms to identify traffic patterns, and generate multi-path traffic feature data; S22. Based on the path status dataset, perform path performance evaluation processing, calculate the performance score of each path using a weighted scoring model, and generate path performance evaluation data. Performance scores are calculated using the following formula: ； in, To score path performance and quantify the quality of a path, , , Let be the weighting coefficient, satisfying , The normalized path available bandwidth. The normalized path delay parameter, This represents the normalized path packet loss rate. S23. The multi-path traffic characteristic data and path performance evaluation data are associated and stored to construct a multi-path performance analysis database.
4. The multi-path cooperative network acceleration method according to claim 1, characterized in that, The step S3, which involves selecting multi-path collaborative strategies based on the multi-path traffic characteristic data and path performance evaluation data to generate multi-path collaborative strategy data, specifically includes the following steps: S31. Obtain the multi-path traffic characteristic data and path performance evaluation data; S32. A multi-objective optimization algorithm is used to select multi-path collaborative strategies, and multi-path collaborative strategy data is generated based on traffic priority and path stability. S33. When the path performance evaluation data is lower than the threshold, the adaptive strategy adjustment mechanism is triggered to regenerate the multi-path collaborative strategy data.
5. The multi-path cooperative network acceleration method according to claim 1, characterized in that, The step S4, which optimizes path resource allocation based on the multi-path coordination strategy data to generate optimized path resource allocation data, specifically includes the following steps: S41. Based on the multi-path collaborative strategy data, perform path resource allocation optimization processing, and use a linear programming model to allocate bandwidth and computing resources. S42. Generate optimized path resource allocation data, including resource quotas and scheduling schedules for each path; S43. Ensure the real-time availability of path resources through the resource reservation protocol, and update and optimize path resource allocation data.
6. The multi-path cooperative network acceleration method according to claim 1, characterized in that, The step S5, which involves performing multi-path data acceleration transmission processing based on the optimized path resource allocation data and generating multi-path accelerated transmission data, specifically includes the following steps: S51. Implement multi-path data acceleration transmission processing based on the optimized path resource allocation data, using data fragmentation and parallel transmission techniques. S52. Perform network packet scheduling processing, and generate multi-path accelerated transmission data based on priority queues and congestion control algorithms; S53. Enhances data transmission security through encryption and compression modules, and accelerates the integrity of transmitted data through real-time verification.
7. The multi-path cooperative network acceleration method according to claim 1, characterized in that, The S6 step involves monitoring and dynamically adjusting the multi-path collaboration effect. Based on the multi-path accelerated transmission data, generating multi-path collaboration monitoring data and dynamic adjustment instructions specifically includes the following steps: S61. Perform multi-path collaboration effect monitoring and processing, and generate multi-path collaboration monitoring data through real-time data stream analysis; S62. Generate dynamic adjustment instructions based on the deviation between monitoring data and preset benchmarks, including path weight adjustment and traffic rerouting instructions; S63. When a path failure is detected, the fault recovery mechanism is automatically triggered to update the multi-path collaborative monitoring data and dynamic adjustment instructions.
8. The multi-path cooperative network acceleration method according to claim 1, characterized in that, The steps in S7 to perform real-time path switching and load balancing based on the multi-path collaborative monitoring data and dynamic adjustment instructions, and to generate the final network acceleration data, specifically include the following: S71. Based on the multi-path collaborative monitoring data and dynamic adjustment instructions, perform real-time path switching processing and use a fast switching algorithm to minimize latency; S72. Implement load balancing to dynamically distribute network traffic to the optimal path and generate final network acceleration data. S73. The acceleration effect analysis is output through the log recording and reporting module, which is used by network administrators for subsequent optimization.
9. The multi-path cooperative network acceleration method according to claim 1, characterized in that, The long-term optimization process of the multi-path cooperative network based on the final network acceleration data in S8 includes historical data analysis and policy base update. The specific steps to generate long-term optimization data are as follows: S81. Regularly collect historical acceleration data to perform trend analysis and pattern recognition; S82. Update the multi-path collaboration strategy library based on the analysis results; S83. Verify the optimization effect through simulation testing to ensure the robustness of the method under different network environments.
10. The multi-path cooperative network acceleration method according to claim 1, characterized in that, The S9 process integrates artificial intelligence-assisted decision processing, and generates intelligent decision data based on the long-term optimized data by using deep learning prediction and reinforcement learning adjustment, specifically including the following steps: S91. Use deep learning models to predict network traffic changes and generate multi-path collaborative strategies in advance; S92. Adaptive acceleration is achieved by dynamically adjusting path selection parameters through reinforcement learning algorithms. S93. Combine AI decision-making results with human intervention.