VPN tunnel communication optimization method based on intelligent routing

Abstract

The invention discloses a VPN (Virtual Private Network) tunnel communication optimization method based on intelligent routing, and particularly relates to the technical field of network communication, which comprises the following steps: S1, sampling multi-dimensional data of each network path according to a preset acquisition period, S2, analyzing parameters adopted in the S1 in real time, and S3, carrying out real-time analysis on the parameters adopted in the S1; s3, comprehensively analyzing the performance indexes of the network paths obtained in the step S2 to obtain comprehensive performance scores of the network paths, and S4, taking the optimal path with the highest score as a VPN tunnel communication path based on the step S3. S5, the VPN tunnel communication is evaluated based on a result obtained after optimization in the step S4, and S6, man-machine interaction is carried out on an evaluation result obtained in the step S5. Through the big data technology, various performance indexes of the network path are comprehensively and dynamically evaluated, the optimal path is intelligently selected according to the actual application requirement, and the data transmission reliability of each network path can be ensured.

Description

Technical Field

[0001] The present invention relates to the field of network communication technologies, and particularly to an optimization method for VPN tunnel communication based on intelligent routing. Background Art

[0002] With the development of digital office and global business, VPN (Virtual Private Network) plays an important role in ensuring network communication security and enabling remote access, and is widely used in places such as enterprise remote office and data encrypted transmission.

[0003] In the field of network communication technologies, traditional VPN routing strategies mainly rely on static or fixed rules or simple algorithms. For example, they select network paths according to a preset priority order, or only consider distance factors. When traditional methods select network paths, they usually only focus on one or a few performance indicators, such as bandwidth or latency, and fail to comprehensively consider various factors affecting VPN communication. In addition to bandwidth and latency, indicators such as packet loss rate, jitter, and network failures also have a significant impact on communication quality.

[0004] Although traditional VPN tunnel communication methods can solve the problem of data transmission, they still have some disadvantages. For example, traditional VPN routing strategies based on simple algorithms cannot real-time perceive the dynamic changes of network status. In the case of network congestion, link failures, etc., they may continuously select paths with poor performance, resulting in high data transmission latency and serious packet loss, thereby reducing the efficiency and quality of VPN communication. For applications with high real-time requirements that only focus on a few performance indicators (such as video conferencing, real-time monitoring, etc.), even if the bandwidth is sufficient, if the packet loss rate is too high or the jitter is too large, the communication quality will also seriously decline. In summary, traditional VPN tunnel communication methods can no longer meet the growing demand for high-quality, secure, and dynamic network communication, and there is an urgent need for an optimization method based on intelligent routing to solve the above problems. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide an optimization method for VPN tunnel communication based on intelligent routing to solve the problems of poor network path selection and lack of comprehensive evaluation of performance indicators proposed in the above background art.

[0006] To achieve the above object, the present invention provides the following technical solution: An optimization method for VPN tunnel communication based on intelligent routing, including:

[0007] S1: Deploy monitoring devices at key nodes of the VPN network, sample multi-dimensional data of each network path according to a preset acquisition period, and obtain various parameters for analyzing the performance of each network path;

[0008] S2: Based on S1, through big data analysis technology, perform real-time analysis on various parameters of S1 to obtain the performance indicators of each network path. The performance indicators of each network path include the performance indicators of each network path status, the encryption performance indicators of each network path, and the fault recovery performance indicators of each network path.

[0009] S3: Through big data analysis technology, comprehensively analyze the performance indicators of each network path obtained in S2 to obtain the comprehensive performance scores of each network path, and make intelligent routing decisions based on the comprehensive performance scores.

[0010] S4: Based on the intelligent routing decision result obtained in S3, use the optimal path as the VPN tunnel communication path, and real-time monitor the performance indicators of the target VPN tunnel communication network path. Trigger the VPN tunnel communication optimization condition according to the abnormal performance indicator monitoring result and perform re-routing optimization.

[0011] S5: Through big data analysis technology, evaluate the VPN tunnel communication based on the result after re-routing optimization in S4 to obtain the VPN tunnel communication optimization evaluation indicators.

[0012] S6: Perform human-computer interaction on the evaluation result of the VPN tunnel communication optimization evaluation indicators obtained in S5.

[0013] Technical effects and advantages of the present invention:

[0014] 1. By deploying monitoring devices at key nodes of the VPN network, the present invention samples multi-dimensional data of each network path according to a preset collection period. This data provides comprehensive network status information for the intelligent routing algorithm; the preset collection period ensures the real-time and continuity of the data, and ensures the stability and efficiency of data transmission.

[0015] 2. Through big data technology, the present invention dynamically evaluates various performance indicators of network paths in real-time and comprehensively, and intelligently selects the optimal path according to actual application requirements, which can ensure the reliability of data transmission on each network path, reduce the risk of failures, significantly improve the efficiency and quality of VPN communication, and meet the strict requirements for network communication in different application scenarios.

[0016] 3. By real-time monitoring the performance indicators of the target VPN tunnel communication network path, triggering the VPN tunnel communication optimization condition according to the abnormal performance indicator monitoring result and performing re-routing optimization, the present invention ensures that the VPN system can quickly switch to a better path in the face of sudden situations such as network path failures and network congestion, enhancing the adaptability of the VPN system to complex and changeable network environments. Description of the Drawings

[0017] Figure 1 It is a schematic diagram of the overall process of the present invention.

[0018] Figure 2 This is a schematic flowchart of the method of the present invention. Specific embodiments

[0019] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0020] Please refer to Figure 1 As shown, the present invention provides an optimization method for VPN tunnel communication based on intelligent routing, including a network path multi-dimensional data acquisition module, a network path performance analysis module, an intelligent routing decision module, a VPN tunnel communication optimization module, a VPN tunnel communication optimization evaluation module, and a VPN tunnel communication optimization human-computer interaction module.

[0021] The network path multi-dimensional data acquisition module is connected to the network path performance analysis module. The intelligent routing decision module is respectively connected to the network path performance analysis module and the VPN tunnel communication optimization module. The VPN tunnel communication optimization module is connected to the network path performance analysis module. The VPN tunnel communication optimization evaluation module is respectively connected to the VPN tunnel communication optimization module and the VPN tunnel communication optimization human-computer interaction module.

[0022] Network path multi-dimensional data acquisition module: used to deploy monitoring devices at key nodes of the VPN network, sample various network path multi-dimensional data according to a preset acquisition period, obtain various parameters for analyzing the performance of each network path, and transmit each parameter to the network path performance analysis module;

[0023] Network path performance analysis module: used to perform real-time analysis on each parameter by the multi-dimensional data acquisition module, obtain the performance indicators of each network path, and transmit the performance indicators to the intelligent routing decision module and the VPN tunnel communication optimization module;

[0024] Intelligent routing decision module: used to comprehensively analyze the performance indicators of each network path obtained by the performance analysis module, obtain the comprehensive performance scores of each network path, make intelligent routing decisions according to the comprehensive performance scores, and transmit the decision results to the VPN tunnel communication optimization module;

[0025] VPN tunnel communication optimization module: dynamically establish a VPN tunnel according to the decision result obtained by the intelligent routing decision module, and monitor the VPN tunnel communication in real time. Trigger the VPN tunnel communication optimization condition according to the monitoring result of the abnormal performance indicator, and transmit the optimized result to the VPN tunnel communication optimization evaluation module;

[0026] VPN Tunnel Communication Optimization Evaluation Module: It is used to evaluate the optimized results obtained by the VPN tunnel communication optimization module and transmit the evaluation results to the VPN tunnel communication optimization human-computer interaction module;

[0027] VPN Tunnel Communication Optimization Human-Computer Interaction Module: It is used to transmit the evaluation results obtained by the evaluation module to the management personnel information terminal for human-computer interaction.

[0028] Please refer to Figure 2 As shown in the figure, a VPN tunnel communication optimization method based on intelligent routing includes the following steps: S1: Deploy monitoring devices at key nodes of the VPN network, sample multi-dimensional data of each network path according to a preset collection period, and obtain various parameters for analyzing the performance of each network path; S2: Based on S1, through big data analysis technology, perform real-time analysis on the various parameters in S1 to obtain the performance indicators of each network path. The performance indicators of each network path include the status performance indicators of each network path, the encryption performance indicators of each network path, and the fault recovery performance indicators of each network path; S3: Through big data analysis technology, comprehensively analyze the performance indicators of each network path obtained in S2 to obtain the comprehensive performance scores of each network path, and make intelligent routing decisions based on the comprehensive performance scores; S4: Based on the intelligent routing decision results obtained in S3, use the optimal path as the VPN tunnel communication path, and real-time monitor the performance indicators of the target VPN tunnel communication network path. Trigger the VPN tunnel communication optimization conditions according to the abnormal performance indicator monitoring results and perform re-routing optimization; S5: Through big data analysis technology, evaluate the VPN tunnel communication based on the results after re-routing optimization in S4 to obtain the VPN tunnel communication optimization evaluation indicators; S6: Perform human-computer interaction on the evaluation results of the VPN tunnel communication optimization evaluation indicators obtained in S5.

[0029] S1: Deploy monitoring devices at key nodes of the VPN network, sample multi-dimensional data of each network path according to a preset collection period, and obtain various parameters for analyzing the performance of each network path. The various parameters include network path status performance parameters, network path encryption performance parameters, and network path fault recovery performance parameters; among them, the network path status performance parameters include bandwidth, delay, packet loss quantity, and queue length; the network path encryption performance parameters include encryption rate and encryption delay, decryption rate and decryption delay, number of times of changing encryption keys, and number of security vulnerabilities; the network path fault recovery performance parameters include average fault detection time and allowed average fault detection time, fault recovery time and allowed fault recovery time, number of fault maintenance window operations, and allowed number of fault maintenance windows;

[0030] It should be specifically noted in this embodiment that the preset acquisition period can be once every 5 seconds or 10 seconds; the key nodes of the VPN include client devices (such as personal computers and mobile terminals), VPN servers, and key routers and switches in the network, etc.; by installing specialized network performance monitoring software on the monitoring device, various parameters of the performance of each network path are collected and analyzed.

[0031] S2: Through big data analysis technology, various parameters in S1 are analyzed in real time to obtain the performance indicators of each network path. The performance indicators of each network path include the performance indicators of the status of each network path, the encryption performance indicators of each network path, and the fault recovery performance indicators of each network path. The real-time analysis of various parameters includes the following steps:

[0032] A1: During the acquisition period, through the monitoring device, the bandwidth bw, delay t_d, number of lost packets n_pl, and queue length ql of each network path are respectively collected; then through big data analysis technology, the network path status performance indicators NSPI are obtained. max(bw) represents the maximum bandwidth value, plr represents the packet loss rate, plr = n_pl / N_tdp, N_tdp represents the amount of data transmitted, dj represents the delay jitter coefficient, dj = σ(t_d) / μ(t_d), σ(t_d) represents the standard deviation of the delay, μ(t_d) represents the average value of the delay, n_tdp represents the amount of data transmitted, the unit of the amount of data transmitted and the bandwidth has been unified, ql represents the queue length, max(ql) represents the maximum queue length, and the queue length refers to the number of data packets in the queue set by the network device to cache and wait for forwarding.

[0033] It should be specifically noted in this embodiment that the unit of the amount of data transmitted and the bandwidth can be bits per second or bytes per second; the SNMP protocol is used to query the interface traffic information of the network device to obtain real-time bandwidth data; the delay is measured by sending an ICMP echo request message to the target node and recording the round-trip time; the sending and receiving situations of data packets are monitored at the network interface, and the number of lost packets is counted to calculate the packet loss rate; for the media stream transmitted using the RTP protocol, the jitter situation is monitored by analyzing the timestamp information in the RTP header; the network congestion degree is evaluated by monitoring the network device queue length and link load situation; by analyzing the network path status performance indicators, low latency and high reliability of each network path can be ensured.

[0034] A2: During the acquisition period, through the monitoring device, the encryption rate v jia and encryption delay t jia of each network path are respectively collected, the decryption rate v jie and decryption delay t jie, the number of times n_ky of replacing the encryption key and the number of security vulnerabilities n_sv; then through big data analysis technology, the network path encryption performance index NPPI is obtained. v jia ref and v jie ref respectively represent the reference values of the encryption rate and the decryption rate, and t jia ref and t jie ref respectively represent the reference values of the encryption delay and the decryption delay. T represents the collection period, and adding 1 to the logarithmic function avoids the value of the logarithmic function ≤ 0.

[0035] It should be specifically stated in this embodiment that for VPN tunnel communication that requires encryption, an advanced encryption algorithm (such as AES, etc.) is adopted to ensure the security of data transmission; during the encryption process, the additional overhead on the data is minimized as much as possible to improve the encryption efficiency; the amount of data processed for encryption / decryption operations is measured through a performance testing tool, and the encryption rate and the decryption rate are obtained by the ratio of the amount of data to the collection period; the encryption delay and the decryption delay are obtained by measuring the round-trip time difference of the data packets before and after encryption / decryption; the number of times of replacing the encryption key is obtained by monitoring the key management log; the number of security vulnerabilities of each network path is obtained through a security scanning tool or a vulnerability database; the data transmission efficiency and security of each network path can be evaluated by analyzing the network path encryption performance index.

[0036] A3: During the collection period, the average fault detection time t_d of each network path, the allowed average fault detection time (t_d)0, the fault recovery time t_r, the allowed fault recovery time (t_r)0, the number of times n_wo of fault maintenance window operations, and the allowed number of times (n_wo)0 of fault maintenance windows are obtained through the fault monitoring log. The number of times of fault maintenance window operations is, for example, restarting the device, updating software, etc.; then through big data analysis technology, the network path fault recovery performance index FRPI is obtained. Δt_d represents the difference between t_d and (t_d)0. If Δt_d is less than 0, then Δt_d is 0. Δt_r represents the difference between t_r and (t_r)0. If Δt_r is less than 0, then Δt_r is 0. Δn_wo represents the difference between n_wo and (n_wo)0. If Δn_wo is less than 0, then Δn_wo is 0.

[0037] In this embodiment, it is necessary to specifically describe the faults of each network path, such as network congestion, high packet loss rate, routing errors, etc.; through the fault monitoring log, the average time from the occurrence of the fault to its detection is recorded to obtain the average fault detection time; the fault handling process is recorded and the time is calculated to obtain the fault recovery time; the maintenance operation log is recorded to obtain the number of fault maintenance window operations; by analyzing the fault recovery performance indicators of each network path, the availability of the path can be evaluated, the vulnerable paths can be identified, and routing selection can be performed, thereby improving the overall network performance.

[0038] S3: Through big data analysis technology, comprehensively analyze the performance indicators of each network path obtained in S2 to obtain the comprehensive performance scores of each network path, and make intelligent routing decisions based on the comprehensive performance scores, including the following steps:

[0039] B1: Through big data analysis technology, comprehensively analyze the performance indicators of each network path to obtain the comprehensive performance score Score of each network path.

[0040] B2: First, through the client of the VPN network, receive the user data transmission request and identify the application type of the data transmission request (such as video conferencing, file transfer, etc.); then, based on the weight database of historical application types, obtain the average value of the weights of the target application type to dynamically allocate the weights of the target application type, and obtain the weights of each performance indicator of the target application type, ω1 + ω2 + ω3 = 1. For example, for video conferencing, the network path status performance indicators and the fault recovery performance indicators of each network path require high weights, ω1 = 0.5, ω2 = 0.2, and ω3 = 0.3.

[0041] B3: Through big data analysis technology, sort according to the comprehensive performance score Score of each network path, and the intelligent routing makes a decision based on the comprehensive performance score, selects the network path with the highest score as the optimal path, and the optimal path information is the intelligent routing decision result.

[0042] In this embodiment, it should be specifically noted that the application type of the data transmission request can be based on the application type set by the user (such as file transfer, video conferencing, online games, etc.) or the system automatically identifies the application type according to the application data characteristics (such as detecting a large amount of real-time media stream data and judging it as a video application), and different weights are dynamically allocated to each performance indicator; for example, path status performance (delay, jitter, packet loss rate): directly affects the user experience and requires high weight; encryption performance: applications with high security requirements (such as finance, medical) require high weight; fault recovery performance: critical services (such as online games) require high weight.

[0043] S4: Based on the intelligent routing decision result obtained in S3, use the optimal path as the VPN tunnel communication path, and monitor the performance metrics of the target VPN tunnel communication network path in real time. Trigger the VPN tunnel communication optimization condition according to the abnormal performance metric monitoring result and perform re-routing optimization. The optimization includes the following steps:

[0044] C1: Through big data analysis technology, based on the target application type, compare the performance metrics of each network path obtained in S2 with the corresponding thresholds respectively. If at least one is less than the threshold, it is determined that the VPN tunnel communication optimization condition is triggered and re-routing optimization is performed; otherwise, if all are greater than or equal to the threshold, it is regarded as normal;

[0045] C2: Transmit the triggered VPN tunnel communication optimization condition to S2 - S3 for re-routing, recalculate the comprehensive performance scores of each network path, re-select the network path with the highest score as the optimal path, and perform path switching for VPN tunnel communication;

[0046] It should be specifically noted in this embodiment that the performance metrics of each network path include the network path status performance index NSPI, the network path encryption performance index NPPI, and the network path fault recovery performance index FRPI of each network path. The corresponding thresholds are NSPI0, NPPI0, and FRPI0 respectively. Compare NSPI with NSPI0, NPPI with NPPI0, and FRPI with FRPI0 respectively to obtain three comparison results. If the number of comparison results less than the threshold ≥ 1, it is determined that the VPN tunnel communication optimization mechanism is triggered; when the NSPI of a certain path > the threshold but the NPPI < the threshold: if it is financial data transmission, immediately switch to the standby path with the highest NPPI; if it is ordinary browsing, continue to use the original path after reducing the encryption strength (AES - 256 → AES - 128).

[0047] It should be specifically noted in this embodiment that for different target application types, the threshold settings of the network path status performance metrics, encryption performance metrics, and fault recovery performance metrics of each network path are different. A unified threshold may cause all applications to trigger alarms due to a single network fluctuation, increasing the operation and maintenance complexity; according to different application types, such as critical services (such as video conferencing), the thresholds are more stringent, and for non-critical services, the thresholds are more relaxed.

[0048] S5: Through big data analysis technology, evaluate the VPN tunnel communication based on the result of re-routing optimization in S4 to obtain the VPN tunnel communication optimization evaluation index VPN_EI, n_rh represents the average number of routing hops, which is obtained by the routing tracing tool to get the average number of routing paths that the data packet passes from the source to the destination. f represents the routing change frequency, which is obtained by the ratio of the number of times the routing path changes to the collection period. μ(dv) and σ(dv) respectively represent the average value and standard deviation of all VPN tunnel communication traffic, which are obtained by analyzing the traffic logs to get the traffic on each VPN tunnel communication to obtain the average value and standard deviation of the traffic. The logarithmic function plus 1 is to avoid the value of the logarithmic function ≤ 0;

[0049] S6: Perform human-computer interaction on the evaluation results of the VPN tunnel communication optimization evaluation metrics obtained in S5. If the VPN tunnel communication optimization evaluation metrics belong to the set allowable range, it indicates that the VPN tunnel communication optimization is good. Otherwise, prompt the management personnel to take measures in a timely manner, such as deploying a traffic shaping strategy to limit the maximum bandwidth of a single user to 50 Mbps; confirm whether the network topology is reasonable to avoid redundant paths; analyze the routing protocol logs to locate abnormal routing updates, etc.

[0050] Secondly: In the accompanying drawings of the disclosed embodiments of the present invention, only the structures related to the disclosed embodiments are involved. For other structures, the common design can be referred to. Without conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0051] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. An optimization method for VPN tunnel communication based on intelligent routing, characterized in that: Including: S1: Deploy monitoring devices at key nodes of the VPN network, sample multi-dimensional data of each network path according to a preset collection period, and obtain various parameters for analyzing the performance of each network path; S2: Based on S1, through big data analysis technology, perform real-time analysis on the various parameters of S1 to obtain the performance indicators of each network path. The performance indicators of each network path include the status performance indicators of each network path, the encryption performance indicators of each network path, and the fault recovery performance indicators of each network path; S3: Through big data analysis technology, comprehensively analyze the performance indicators of each network path obtained in S2 to obtain the comprehensive performance scores of each network path, and make intelligent routing decisions according to the comprehensive performance scores; S4: Based on the intelligent routing decision result obtained in S3, use the optimal path as the VPN tunnel communication path, and real-time monitor the performance indicators of the target VPN tunnel communication network path. Trigger the VPN tunnel communication optimization condition according to the abnormal performance indicator monitoring result and perform re-routing optimization; S5: Through big data analysis technology, evaluate the VPN tunnel communication based on the result after re-routing optimization in S4 to obtain the VPN tunnel communication optimization evaluation index; S6: Perform human-computer interaction on the evaluation result VPN tunnel communication optimization evaluation index obtained in S5.

2. The optimized method for VPN tunnel communication based on intelligent routing according to claim 1, characterized in that: The various parameters in S1 include network path status performance parameters, network path encryption performance parameters, and network path fault recovery performance parameters; Among them, the network path status performance parameters include bandwidth, delay, packet loss quantity, and queue length; the network path encryption performance parameters include encryption rate and encryption delay, decryption rate and decryption delay, number of times of changing encryption keys, and number of security vulnerabilities; The network path fault recovery performance parameters include average fault detection time and allowable average fault detection time, fault recovery time and allowable fault recovery time, number of fault maintenance window operations and allowable number of fault maintenance windows.

3. A method for optimizing VPN tunnel communication based on intelligent routing according to claim 1, characterized in that: Obtain the performance indicators of each network path status in S2: During the collection period, through the monitoring device, collect the bandwidth bw, delay t_d, number of lost packets n_pl, and queue length ql of each network path respectively; then through big data analysis technology, obtain the network path status performance indicator NSPI, max(bw) represents the maximum bandwidth value, plr represents the packet loss rate, plr = n_pl / N_tdp, N_tdp represents the amount of data transmitted, dj represents the delay jitter coefficient, dj = σ(t_d) / μ(t_d), σ(t_d) represents the standard deviation of the delay, μ(t_d) represents the average value of the delay, n_tdp represents the amount of data transmitted, the unit of the amount of data transmitted and the bandwidth has been unified, ql represents the queue length, max(ql) represents the maximum queue length, and the queue length refers to the number of data packets in the queue set by the network device to cache and wait for forwarding.

4. An optimization method for VPN tunnel communication based on intelligent routing according to claim 1, characterized in that: The encryption performance metrics of each network path obtained in S2 are as follows: during the collection period, the encryption rate v of each network path is collected separately by the monitoring device jia and the encryption delay t jia , the decryption rate v jie and the decryption delay t jie , the number of times n_ky of replacing the encryption key, and the number of security vulnerabilities n_sv; then, through big data analysis technology, the encryption performance metrics NPPI of each network path are obtained v jia ref and v jie ref respectively represent the reference values of the encryption rate and the decryption rate, and t jia ref and t jie ref respectively represent the reference values of the encryption delay and the decryption delay. T represents the acquisition period, and adding 1 to the logarithmic function avoids the value of the logarithmic function being ≤ 0.

5. A method for optimizing VPN tunnel communication based on intelligent routing according to claim 1, characterized in that: The fault recovery performance indicators of each network path obtained in S2: During the collection period, obtain the average fault detection time t_d and allowable average fault detection time (t_d)0, fault recovery time t_r and allowable fault recovery time (t_r)0, number of fault maintenance window operations n_wo and allowable number of fault maintenance windows (n_wo)0 of each network path through the fault monitoring log; then through big data analysis technology, obtain the fault recovery performance indicator FRPI of each network path, Δt_d represents the difference between t_d and (t_d)0. If Δt_d is less than 0, then Δt_d is 0. Δt_r represents the difference between t_r and (t_r)0. If Δt_r is less than 0, then Δt_r is 0. Δn_wo represents the difference between n_wo and (n_wo)0. If Δn_wo is less than 0, then Δn_wo is 0.

6. The optimized method for VPN tunnel communication based on intelligent routing according to claim 1, characterized in that: The intelligent routing decision in S3 includes the following steps: A1: Through big data analysis technology, comprehensively analyze the performance indicators of each network path to obtain the comprehensive performance score Score of each network path. A2: First, through the client of the VPN network, receive the user data transmission request and identify the application type of the data transmission request; then based on the weight database of historical application types, obtain the average value of the weights of the target application type to dynamically allocate the weights of the target application type, and obtain the weights of the performance indicators of the target application type, ω1 + ω2 + ω3 = 1; A3: Through big data analysis technology, sort according to the comprehensive performance scores Score of each network path. The intelligent router makes a decision based on the comprehensive performance scores and selects the network path with the highest score as the optimal path. The optimal path information is the intelligent routing decision result.

7. A method for optimizing VPN tunnel communication based on intelligent routing according to claim 1, characterized in that: The optimization in S4 includes the following steps: B1: Through big data analysis technology, based on the target application type, compare the performance indicators of each network path obtained in S2 with the corresponding thresholds respectively. If at least one is less than the threshold, it is determined that the VPN tunnel communication optimization condition is triggered and re-routing optimization is performed; otherwise, if all are greater than or equal to the threshold, it is regarded as normal. B2: Transmit the triggered VPN tunnel communication optimization condition to S2 - S3 for re-routing, recalculate the comprehensive performance scores of each network path, re-select the network path with the highest score as the optimal path, and perform path switching for the VPN tunnel communication.

8. A method for optimizing VPN tunnel communication based on intelligent routing according to claim 1, characterized in that: Obtain the VPN tunnel communication optimization evaluation index VPN_EI in S5. n_rh represents the average number of routing hops, and n_rh is obtained by using a routing trace tool to obtain the average number of routing paths that a data packet passes through from the source to the destination. f represents the routing change frequency, which is obtained by the ratio of the number of routing path changes to the collection period. μ(dv) and σ(dv) respectively represent the average value and standard deviation of all VPN tunnel communication traffic. The average value and standard deviation of the traffic on each VPN tunnel communication are obtained by analyzing the traffic logs. Adding 1 to the logarithmic function avoids the value of the logarithmic function being ≤ 0.

Images