Wireless communication network traffic control system under edge computing gateway
By designing a wireless communication network traffic control system in an edge computing gateway, and evaluating and adjusting traffic shaping strategies using neural network algorithms, the problems of network congestion and data packet loss are solved, and the efficient and reliable operation of the wireless communication network is achieved.
Patent Information
- Application Number
- CN202411245324.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-06
- Publication Date
- 2024-12-24
- Estimated Expiration
- 2044-09-06
AI Technical Summary
During the traffic shaping process, existing edge computing gateways have hidden dangers of network congestion and data packet loss due to unreasonable buffer allocation or burst traffic shocks.
A wireless communication network traffic control system under an edge computing gateway was designed, including data acquisition module, hidden danger assessment module, maintenance adjustment module and traffic control module. The traffic plastic shaping potential hazard assessment model was established through neural network algorithms, and the traffic packet loss was monitored in real time. Sculpting delay and plastic shaping event exceptions, dynamically adjust the buffer size and thread bandwidth proportion, and optimize traffic control strategies.
It effectively reduces the risks of network congestion and data packet loss, improves network stability and reliability, and ensures efficient operation under various loads.
Smart Images
Figure CN118764899B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of edge computing gateways, and more specifically, to a wireless communication network traffic control system under an edge computing gateway. Background Art
[0002] Edge computing gateway refers to a computing device or service located at the edge of the network, which is used to process and manage data and computing tasks directly connected to edge devices such as sensors and terminal devices to reduce latency and improve efficiency.
[0003] A wireless communication network is a network system that uses wireless technology to transmit information. It is widely used in various scenarios, from personal devices to large-scale enterprises and urban infrastructure. These networks use wireless transmission media such as radio waves or infrared rays to achieve the transmission and exchange of data, voice and video. Wireless communication networks can include a variety of technologies and protocols, such as Wi-Fi, Bluetooth, Zigbee, LTE (Long-Term Evolution) and 5G, each of which has its own specific transmission speed, coverage and power consumption requirements. The main advantages of wireless communication networks include convenience, flexibility and scalability, which make them widely used in mobile communications, the Internet of Things, smart cities, industrial automation and other fields. With the advancement of technology and the growth of demand, wireless communication networks are constantly evolving and expanding to meet increasingly complex communication needs and application scenarios.
[0004] A traffic control system is a system used to manage and optimize data flow, aiming to ensure efficient transmission and processing of data in the network; these systems achieve their functions by monitoring data traffic, adjusting transmission rates, handling congestion, and optimizing route selection; traffic control systems usually include various algorithms and strategies, such as congestion control algorithms (such as TCP congestion control), traffic scheduling algorithms (such as QoS), packet scheduling algorithms (such as WFQ), routing optimization algorithms, etc.; its main goal is to improve network performance and efficiency, reduce delays and packet loss rates, and ensure quality of service (QoS), especially in the face of large-scale data transmission, multimedia streaming and real-time applications; the design and implementation of traffic control systems is crucial to the reliability and manageability of modern network infrastructure, and can effectively cope with the growing data traffic and diverse application requirements.
[0005] The token bucket algorithm is one of the common techniques for traffic shaping. It uses a "bucket" to store tokens, and each token represents a data packet or byte that is allowed to be sent. Tokens are added to the bucket at a fixed rate, and the capacity of the bucket is limited. Tokens that exceed the capacity will be discarded.
[0006] The leaky bucket algorithm shapes traffic by simulating a "leaky bucket" with a fixed capacity. After a data packet enters the leaky bucket, it "leaks" from the bucket (i.e. is sent to the network) at a fixed rate. If the rate at which data packets arrive exceeds the "leakage" rate of the leaky bucket, the excess data packets will be discarded.
[0007] For example, the invention patent with announcement number CN114143836B announces a load balancing method, device and terminal based on edge computing gateway, including: obtaining the working parameters of the target edge computing gateway; determining the load of the target edge computing gateway based on the working parameters and the preset load evaluation rules; determining the priority of the target edge computing gateway according to the load of the target edge computing gateway and the weight of the target edge computing gateway; determining the balanced edge computing gateway corresponding to the target edge computing gateway among all edge computing gateways based on the priority of the target edge computing gateway and the preset fold balancing rule; when the load of the target edge computing gateway is greater than the preset load, controlling the target edge computing gateway and the balanced edge computing gateway to jointly process the computing tasks of the target edge computing gateway. The present invention can balance the computing resources of the edge computing gateway, enable the edge computing gateway to efficiently calculate, and thus improve the timeliness of data transmission.
[0008] For example, the invention patent with announcement number: CN113727434B announces a vehicle-road cooperative auxiliary positioning system and method based on edge computing gateway, the positioning system includes a vehicle terminal, an edge computing gateway, a data acquisition unit and a cloud server; the data acquisition unit is used to collect road information of the road; the edge computing gateway is used to send the road information to the cloud server; the cloud server is used to generate a road model based on the road information, and send the road model to each edge computing gateway; the edge computing gateway is also used to determine the location information of the calibrated vehicle according to the signal strength value of the V2X communication between each edge computing gateway and the vehicle terminal in the detection interval and the gateway coordinates of each edge computing gateway in the detection interval when the calibrated vehicle is traveling in the detection interval of the road. In this way, even if the vehicle travels on a road with no satellite signal or poor GPS signal, it can still accurately locate the unknown vehicle on the road.
[0009] The above-disclosed technical solution has at least the following technical problems: the existing edge computing gateway sets different service quality policies through QoS management. The edge computing gateway can prioritize data flows according to application requirements to control traffic. However, the existing method used for traffic shaping in the core step of QoS management is usually a leaky bucket algorithm or a token bucket algorithm; however, although traffic shaping is intended to smooth traffic, due to unreasonable buffer allocation or sudden traffic impact, the wireless communication network may have the risk of network congestion and data packet loss.
[0010] In view of the above problems, the present invention proposes a solution. Summary of the invention
[0011] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a wireless communication network traffic control system under an edge computing gateway, which solves the hidden dangers of network congestion and data packet loss in the wireless communication network due to unreasonable buffer allocation or sudden traffic impact through wireless communication network traffic control.
[0012] To achieve the above object, the present invention provides the following technical solutions:
[0013] The wireless communication network traffic control system under the edge computing gateway includes a data acquisition module, a hidden danger assessment module, a maintenance adjustment module and a traffic control module, and there are connections between the modules; the data acquisition module is used to obtain the traffic data and shaping algorithm data of the edge computing gateway for traffic shaping at the same time interval within a period of time; the hidden danger assessment module is used to establish a traffic shaping hidden danger assessment model based on the traffic data and the shaping algorithm data based on a neural network algorithm, and generate a traffic shaping hidden danger assessment coefficient; the maintenance adjustment module is used to perform maintenance and adjustment on the buffer size and each thread bandwidth proportion of the shaping algorithm based on the traffic shaping hidden danger assessment coefficient; the traffic control module is used to obtain several traffic shaping hidden danger assessment coefficients of the communication network at the same time interval during the maintenance and adjustment process, construct an adjustment set, and draw a three-dimensional adjustment curve based on the adjustment set, and obtain the optimal buffer size and each thread bandwidth proportion through curve analysis.
[0014] In a preferred embodiment, the traffic data includes a traffic packet loss anomaly impact coefficient; the shaping algorithm data includes a shaping delay impact coefficient and a shaping event anomaly coefficient.
[0015] In a preferred embodiment, the specific method for obtaining the traffic packet loss anomaly impact coefficient is as follows: in a wireless communication network, data packet interaction data of each time period is collected, including timestamps, data packet sizes, and packet loss events; based on the data packet interaction data and based on the pine-cypress distribution, the probability distribution of packet loss events is fitted, and the average packet loss rate is calculated; the actual packet loss rate is obtained, and the expected value and variance of the packet loss rate are calculated; the packet loss rate anomaly assessment value is obtained by comparing and calculating the difference between the actual packet loss rate and the expected value; the correlation between the packet loss rate and the network load and bandwidth utilization rate is calculated based on the Pearson correlation coefficient to obtain the packet loss rate impact coefficient; the probability density function of the packet loss rate is calculated based on the kernel density estimation as the distribution feature of the packet loss rate; the traffic packet loss anomaly impact coefficient is calculated based on the distribution feature of the packet loss rate combined with the packet loss rate impact coefficient and the packet loss rate anomaly assessment value.
[0016] In a preferred embodiment, the specific method for obtaining the shaping delay impact coefficient is as follows: real-time monitoring and recording of the delay time of each data packet before and after being processed by the shaping algorithm; calculating the delay difference of each data packet based on the delay time before and after processing, and constructing a delay difference sequence; recording the variation range of the delay difference sequence within a period of time, and calculating the volatility of the delay difference; summarizing the cumulative delay differences of the shaping algorithm within a period of time to obtain a cumulative delay effect; defining a threshold for delay-triggered congestion based on historical data; when the cumulative delay effect exceeds the threshold for delay-triggered congestion, marking the current time period as a potential congestion period; performing time series segmentation on the historical shaping data to obtain a number of time series data, and determining whether they are potential congestion periods, respectively, as historical congestion events; mapping the time series data with historical congestion events based on random forests to generate a probability model for delay-triggered congestion; calculating the shaping delay impact coefficient based on the probability model for delay-triggered congestion and the volatility of the delay difference.
[0017] In a preferred embodiment, the specific method for obtaining the shaping event anomaly coefficient is as follows: based on historical traffic shaping event types, a traffic shaping event set is obtained, and the traffic shaping event set includes various traffic shaping algorithms and their traffic shaping events and event effects; various traffic shaping event types in the traffic shaping event set are assigned preset anomaly scores; a regression model is constructed based on the traffic shaping event set and the corresponding anomaly scores to generate a shaping event anomaly coefficient.
[0018] In a preferred embodiment, the buffer size of the shaping algorithm and the bandwidth share of each thread are maintained and adjusted based on the traffic shaping risk assessment coefficient. Specifically, when the traffic shaping risk assessment coefficient is greater than a preset maintenance threshold, a maintenance request is issued, the buffer size is adjusted arithmetic based on a preset iteration step, and an internal random adjustment is made to the bandwidth share of each thread with a sum of 1.
[0019] In a preferred embodiment, the adjustment set draws a three-dimensional adjustment curve, specifically: a three-dimensional rectangular coordinate system is established with the bandwidth ratio and buffer size of each thread as the x-axis and y-axis, and the traffic shaping risk assessment coefficient as the z-axis; a three-dimensional curve graph is drawn according to the traffic shaping risk assessment coefficient and the corresponding bandwidth ratio and buffer size of each thread, the bandwidth ratio and buffer size of each thread at the lowest point in the curve graph are taken as the optimal buffer size and bandwidth ratio of each thread, and the traffic control of the wireless communication network is adjusted.
[0020] Technical effects and advantages of the wireless communication network traffic control system under the edge computing gateway of the present invention:
[0021] 1. The present invention realizes refined management of wireless network traffic through the effective integration of data acquisition module, hidden danger assessment module, maintenance adjustment module and traffic control module; the system can monitor key indicators such as traffic packet loss, shaping delay and shaping event anomaly in real time, provide timely warning, accurately locate the source of network problems, and optimize traffic control strategy, thereby improving the stability and reliability of the network; by analyzing these anomaly coefficients, the system not only effectively reduces the risk of network congestion and data packet loss, but also supports dynamic optimization of network performance to ensure efficient operation of the network under various loads.
[0022] 2. The present invention establishes a traffic shaping hidden danger assessment model based on a neural network algorithm, comprehensively analyzes the traffic packet loss abnormality impact coefficient, the shaping delay impact coefficient and the shaping event abnormality coefficient, and generates a traffic shaping hidden danger assessment coefficient; it is used to dynamically adjust the buffer size and thread bandwidth ratio of the shaping algorithm to optimize the network traffic control strategy; the method can identify and evaluate network hidden dangers in real time, automatically adjust traffic management parameters, improve network stability and performance, and determine the optimal configuration through three-dimensional adjustment curve analysis to ensure efficient and reliable operation of the wireless communication network. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 It is a schematic diagram of the structure of the wireless communication network traffic control system under the edge computing gateway of the present invention. DETAILED DESCRIPTION
[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0025] Embodiment 1, Figure 1 A wireless communication network traffic control system under the edge computing gateway of the present invention is provided, including a data acquisition module, a hidden danger assessment module, a maintenance adjustment module and a traffic control module, and there are connections between the modules;
[0026] The data acquisition module is used to obtain the traffic data and shaping algorithm data of the edge computing gateway performing traffic shaping at the same time interval within a period of time.
[0027] In this embodiment, the traffic data includes a traffic packet loss anomaly impact coefficient; the shaping algorithm data includes a shaping delay impact coefficient and a shaping event anomaly coefficient.
[0028] The traffic packet loss anomaly impact coefficient is a key indicator used to evaluate network congestion and data packet loss risks in wireless communication networks. It is determined by analyzing the frequency and severity of packet loss events in traffic data, as well as the correlation between packet loss and network congestion. This coefficient can reflect the stability of the network under high load or burst traffic conditions, and help identify and quantify the potential negative impact of traffic shaping strategies on network performance. When the traffic packet loss anomaly impact coefficient increases, it means that the network may have a high risk of congestion, and the traffic control strategy needs to be adjusted to reduce packet loss and ensure the reliability of data transmission and service quality.
[0029] Analyzing the impact coefficient of abnormal traffic packet loss has the following advantages for discovering the hidden dangers of network congestion and data packet loss in wireless communication networks:
[0030] Real-time monitoring and early warning: The impact coefficient of abnormal traffic packet loss can reflect the packet loss and congestion level in the network in real time. By continuously monitoring this coefficient, potential network risks can be discovered early to avoid the accumulation and deterioration of problems.
[0031] Accurately locate the source of the problem: This coefficient can help network managers accurately locate the specific time period or network node that causes packet loss and congestion by quantitatively analyzing the frequency and severity of packet loss events, so as to quickly take targeted optimization measures;
[0032] Optimize traffic control strategies: By analyzing the impact coefficient of abnormal traffic packet loss, the effectiveness of existing traffic shaping and control strategies can be evaluated. If the coefficient is found to be too high, it means that the existing strategy may be insufficient and needs to be adjusted or optimized to improve network stability and reliability.
[0033] Improve network reliability: Regularly analyzing the impact coefficient of abnormal traffic packet loss can help identify weak links in the network and reduce data packet loss by optimizing traffic management measures, thereby improving overall network reliability and user experience.
[0034] Data-driven decision support: This coefficient provides a quantitative indicator that enables network managers to make more scientific and effective decisions based on data, avoiding misjudgments or omissions caused by subjective judgments, and helping to achieve more refined network management;
[0035] With these advantages, the traffic packet loss anomaly impact coefficient becomes an important tool for monitoring and optimizing traffic management in wireless communication networks, helping to promptly detect and resolve network congestion and data packet loss risks, and ensure the smooth operation of the network.
[0036] The specific method for obtaining the traffic packet loss abnormality impact coefficient is as follows:
[0037] In wireless communication networks, data packet interaction data is collected for each time period, including timestamps, packet sizes, and packet loss events;
[0038] Based on the data packet interaction data and the pine-cypress distribution, the probability distribution of packet loss events is fitted and the average packet loss rate is calculated;
[0039] Get the actual packet loss rate and calculate the expected value and variance of the packet loss rate;
[0040] By comparing and calculating the difference between the actual packet loss rate and the expected value, the packet loss rate abnormality assessment value is obtained;
[0041] The correlation between packet loss rate and network load and bandwidth utilization is calculated based on the Pearson correlation coefficient to obtain the packet loss rate impact coefficient;
[0042] The probability density function of the packet loss rate is calculated based on the kernel density estimation as the distribution characteristic of the packet loss rate;
[0043] Based on the distribution characteristics of the packet loss rate, combined with the packet loss rate impact coefficient and the packet loss rate anomaly assessment value, the traffic packet loss anomaly impact coefficient is calculated.
[0044] The specific calculation formula of the packet loss rate abnormal evaluation value is as follows:
[0045]
[0046] The specific calculation formula of the packet loss rate influence coefficient is as follows:
[0047]
[0048] The specific calculation formula of the distribution characteristics of the packet loss rate is as follows:
[0049]
[0050] The specific calculation formula of the traffic packet loss abnormality impact coefficient is as follows:
[0051]
[0052] In the formula, is the packet loss rate abnormal evaluation value, is the actual packet loss rate, is the expected value of packet loss rate, is the packet loss rate variance, is the packet loss rate influence coefficient, is the covariance of packet loss rate and network load, is the standard deviation of the packet loss rate, is the standard deviation of the network load, is the total number of recorded packet loss rate data, is the bandwidth parameter, is the kernel function, is the pth packet loss rate, is the impact coefficient of abnormal traffic packet loss.
[0053] The shaping delay impact coefficient is an important indicator for evaluating the impact of the delay introduced by the traffic shaping algorithm on network performance and stability in wireless communication networks. The traffic shaping algorithm smoothes the traffic by adjusting the sending rate of data packets to reduce the impact of burst traffic on the network. However, in this process, the transmission of data packets may be delayed due to the delay of the shaping algorithm, especially in the case of high load or burst traffic, the delay may further increase the risk of network congestion and data packet loss. The shaping delay impact coefficient helps evaluate the impact of delay on the overall network service quality by quantifying the correlation between the delay introduced by the shaping algorithm and network congestion and packet loss events. The higher the coefficient, the greater the potential negative impact of the shaping delay on network stability and data transmission reliability. Therefore, monitoring and optimizing the shaping delay impact coefficient is crucial to improving the overall performance of wireless communication networks and reducing network congestion and data packet loss risks.
[0054] Analyzing the shaping delay impact coefficient has the following advantages for discovering the hidden dangers of network congestion and data packet loss in wireless communication networks:
[0055] Early warning of network performance degradation: The shaping delay impact coefficient can effectively reflect the potential impact of the delay introduced by the traffic shaping algorithm on the overall network performance. When the coefficient starts to increase, it can provide early warning of possible network congestion problems, so that optimization measures can be taken in time to avoid significant degradation of network performance.
[0056] Optimize traffic shaping strategies: By analyzing the shaping delay impact coefficient, it can be found whether the delay introduced by the existing traffic shaping strategy under specific network conditions exceeds the reasonable range; this provides data support for further optimization and adjustment of the shaping algorithm to ensure more accurate and efficient network traffic management;
[0057] Reduce the risk of packet loss: In the case of burst traffic, excessive shaping delay may lead to packet backlog and loss. The analysis of shaping delay impact coefficient can help identify this potential risk of packet loss, ensure smooth transmission of data packets, and improve network reliability and service quality.
[0058] Improve user experience: By monitoring and analyzing the shaping delay impact coefficient, you can identify delay issues that may cause a decline in user experience. Timely adjustment of shaping strategies can help reduce delays and improve the network experience of end users, especially in application scenarios with high real-time requirements such as video streaming and online games.
[0059] Support dynamic network optimization: The shaping delay impact coefficient provides a dynamic monitoring indicator that can reflect the performance of the traffic shaping algorithm in actual network operation in real time. This provides a basis for implementing dynamic network optimization strategies, enabling the network management system to adjust the shaping parameters in real time according to the current network status to ensure that the network is always in the best operating state.
[0060] The specific method for obtaining the shaping delay influence coefficient is as follows:
[0061] Real-time monitoring and recording of the delay time of each data packet before and after being processed by the shaping algorithm;
[0062] Calculate the delay difference of each data packet according to the delay time before and after processing, and construct a delay difference sequence;
[0063] Record the variation range of the delay difference sequence within a period of time and calculate the volatility of the delay difference;
[0064] Summarize the cumulative delay differences of the shaping algorithm over a period of time to obtain the cumulative delay effect;
[0065] Based on historical data, define the threshold for delay triggering congestion;
[0066] When the cumulative delay effect exceeds the threshold of delay triggering congestion, the current time period is marked as a potential congestion period;
[0067] Perform time series segmentation on the historical shaping data to obtain several time series data, and determine whether they are potential congestion periods, as historical congestion events;
[0068] Mapping time series data with historical congestion events based on random forests to generate a probability model of delay-triggered congestion;
[0069] According to the probability model of delay triggering congestion and the volatility of the delay difference, the shaping delay impact coefficient is calculated.
[0070] The specific calculation formula of the volatility of the delay difference is as follows:
[0071]
[0072] The specific calculation formula of the cumulative delay effect is as follows:
[0073]
[0074] The specific calculation formula of the shaping delay influence coefficient is as follows:
[0075]
[0076] In the formula, is the volatility of the delay difference, is the rth delay difference, is the maximum value in the delay difference sequence, is the minimum value in the delay difference sequence, is the expected delay difference, is the delay difference, is the cumulative delay effect of the vth time series data, is the total number of data in the delay difference sequence, is the shaping delay influence coefficient, Probabilistic model for delay-triggered congestion.
[0077] The shaping event anomaly coefficient is a key indicator to measure the abnormal behavior of the shaping algorithm under specific network conditions. This coefficient evaluates the stability of the algorithm in flow control by monitoring the frequency, duration and impact range of shaping events. When the shaping event anomaly coefficient is high, it often means that the network may have frequent burst traffic or unforeseen delays during the shaping process. These abnormal situations may lead to increased network congestion and increased data packet loss rate, thus posing a potential hidden danger to the overall performance of the wireless communication network.
[0078] Analyzing the abnormal coefficient of shaping events has the following advantages for discovering the hidden dangers of network congestion and data packet loss in wireless communication networks:
[0079] Early warning capability: The shaping event anomaly coefficient can timely capture abnormal behaviors in the operation of the shaping algorithm, such as sudden shaping events or frequent traffic fluctuations, provide early warnings, and help identify potential network congestion and data packet loss risks;
[0080] Refined analysis: This coefficient provides a refined analysis tool by quantifying the degree of abnormality of the shaping event, allowing network managers to distinguish between normal traffic shaping behavior and abnormal behavior, thereby more accurately locating the source of the problem;
[0081] Dynamic adaptability: The shaping event anomaly coefficient can be dynamically adjusted as the network environment and load change, adapting to different network conditions and enhancing the ability to identify hidden dangers in complex network environments;
[0082] Comprehensive risk assessment: The shaping event anomaly coefficient combines data from multiple dimensions, such as event frequency, duration, and impact range, to provide a comprehensive risk assessment method that can comprehensively measure the impact of shaping algorithms on network performance and detect and respond to potential network problems in advance;
[0083] With these advantages, the shaping event anomaly coefficient plays a key role in network performance management, helping to identify and prevent potential network congestion and data packet loss caused by shaping algorithm anomalies.
[0084] The specific method for obtaining the abnormal coefficient of the shaping event is as follows:
[0085] Based on the historical traffic shaping event types, a traffic shaping event set is obtained, wherein the traffic shaping event set includes various traffic shaping algorithms and their traffic shaping events and event effects;
[0086] Assigning a preset anomaly score to each traffic shaping event type in the traffic shaping event set;
[0087] A regression model is constructed based on the traffic shaping event set and the corresponding anomaly scores to generate the shaping event anomaly coefficient.
[0088] The specific calculation formula of the abnormal coefficient of the shaping event is as follows:
[0089]
[0090] In the formula, is the shaping event abnormality coefficient, … is the preset anomaly score corresponding to the traffic shaping event, … is the proportion of traffic shaping events in all traffic shaping events, is the type of traffic shaping event, n is the number of the traffic shaping event, and e is a natural number.
[0091] This embodiment realizes refined management of wireless network traffic through the effective integration of data acquisition module, hidden danger assessment module, maintenance adjustment module and traffic control module; the system can monitor key indicators such as traffic packet loss, shaping delay and shaping event anomalies in real time, provide timely warnings, accurately locate the source of network problems, and optimize traffic control strategies, thereby improving network stability and reliability; by analyzing these anomaly coefficients, the system not only effectively reduces the risk of network congestion and data packet loss, but also supports dynamic optimization of network performance to ensure efficient operation of the network under various loads.
[0092] Embodiment 2, hidden danger assessment module, is used to establish a traffic shaping hidden danger assessment model based on a neural network algorithm based on traffic data and shaping algorithm data, and generate a traffic shaping hidden danger assessment coefficient.
[0093] In this embodiment, the traffic shaping potential risk assessment model is established based on the traffic data and the shaping algorithm data based on the neural network algorithm to generate the traffic shaping potential risk assessment coefficient, which is specifically:
[0094] The acquired traffic packet loss abnormal impact coefficient, shaping delay impact coefficient and shaping event abnormal coefficient are used to establish a traffic shaping potential risk assessment model to generate a traffic shaping potential risk assessment coefficient;
[0095] The specific calculation formula of the traffic shaping hidden danger assessment coefficient is as follows:
[0096]
[0097] In the formula, is the traffic shaping risk assessment coefficient, is the preset proportional coefficient of the traffic packet loss anomaly impact coefficient, The preset proportionality factor for shaping the delay influence factor, is the preset proportional coefficient of the shaping event abnormality coefficient, is the impact coefficient of abnormal traffic packet loss, The preset proportionality factor for shaping the delay influence factor, It is the preset proportional coefficient of the abnormal coefficient of the shaping event.
[0098] The maintenance and adjustment module is used to maintain and adjust the buffer size of the shaping algorithm and the bandwidth proportion of each thread based on the traffic shaping risk assessment coefficient.
[0099] The buffer size of the shaping algorithm and the bandwidth proportion of each thread are maintained and adjusted based on the traffic shaping risk assessment coefficient, specifically:
[0100] When the traffic shaping risk assessment coefficient is greater than the pre-examination maintenance threshold, a maintenance request is issued, the buffer size is adjusted arithmetic based on the preset iteration step, and the bandwidth proportion of each thread is internally randomly adjusted to a sum of 1.
[0101] The traffic control module is used to obtain several traffic shaping risk assessment coefficients of the communication network at the same time interval during the maintenance adjustment process, construct an adjustment set, draw a three-dimensional adjustment curve based on the adjustment set, and obtain the optimal buffer size and bandwidth proportion of each thread through curve analysis.
[0102] The adjustment set draws a three-dimensional adjustment curve, specifically:
[0103] A three-dimensional rectangular coordinate system is established with the bandwidth ratio of each thread and the buffer size as the x-axis and y-axis, and the traffic shaping risk assessment coefficient as the z-axis;
[0104] A three-dimensional curve graph is drawn according to the traffic shaping risk assessment coefficient and the corresponding bandwidth ratio of each thread and buffer size. The bandwidth ratio of each thread and buffer size at the lowest point in the curve graph are taken as the optimal buffer size and bandwidth ratio of each thread, and the traffic control of the wireless communication network is adjusted.
[0105] This embodiment establishes a traffic shaping risk assessment model based on a neural network algorithm, comprehensively analyzes the traffic packet loss abnormality impact coefficient, the shaping delay impact coefficient and the shaping event abnormality coefficient, and generates a traffic shaping risk assessment coefficient; it is used to dynamically adjust the buffer size and thread bandwidth ratio of the shaping algorithm to optimize the network traffic control strategy; the method can identify and evaluate network risks in real time, automatically adjust traffic management parameters, improve network stability and performance, and determine the optimal configuration through three-dimensional adjustment curve analysis to ensure efficient and reliable operation of the wireless communication network.
[0106] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.
[0107] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.
[0108] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0109] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.
[0110] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
[0111] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A wireless communication network traffic control system under an edge computing gateway, characterized in that: It includes a data acquisition module, a hidden danger assessment module, a maintenance adjustment module and a flow control module, and there are connections between the modules; A data acquisition module is used to obtain traffic data and shaping algorithm data of the edge computing gateway performing traffic shaping at the same time interval within a period of time; The traffic data includes a traffic packet loss anomaly impact coefficient; the shaping algorithm data includes a shaping delay impact coefficient and a shaping event anomaly coefficient; The specific method for obtaining the abnormal coefficient of the shaping event is as follows: Based on the historical traffic shaping event types, a traffic shaping event set is obtained, wherein the traffic shaping event set includes various traffic shaping algorithms and their traffic shaping events and event effects; Assigning a preset anomaly score to each traffic shaping event type in the traffic shaping event set; A regression model is constructed based on the traffic shaping event set and the corresponding anomaly scores to generate the shaping event anomaly coefficient; A hidden danger assessment module is used to establish a traffic shaping hidden danger assessment model based on a neural network algorithm based on traffic data and shaping algorithm data, and generate a traffic shaping hidden danger assessment coefficient; A maintenance and adjustment module is used to maintain and adjust the buffer size of the shaping algorithm and the bandwidth ratio of each thread based on the traffic shaping risk assessment coefficient; The traffic control module is used to obtain several traffic shaping risk assessment coefficients of the communication network at the same time interval during the maintenance adjustment process, construct an adjustment set, draw a three-dimensional adjustment curve based on the adjustment set, and obtain the optimal buffer size and bandwidth proportion of each thread through curve analysis.
2. The wireless communication network traffic control system under the edge computing gateway according to claim 1, characterized in that: The specific method for obtaining the traffic packet loss abnormality impact coefficient is as follows: In wireless communication networks, data packet interaction data is collected for each time period, including timestamps, packet sizes, and packet loss events; Based on the data packet interaction data and the pine-cypress distribution, the probability distribution of packet loss events is fitted and the average packet loss rate is calculated; Get the actual packet loss rate and calculate the expected value and variance of the packet loss rate; By comparing and calculating the difference between the actual packet loss rate and the expected value, the packet loss rate abnormality assessment value is obtained; The correlation between packet loss rate and network load and bandwidth utilization is calculated based on the Pearson correlation coefficient to obtain the packet loss rate impact coefficient; The probability density function of the packet loss rate is calculated based on the kernel density estimation as the distribution characteristic of the packet loss rate; Based on the distribution characteristics of the packet loss rate, combined with the packet loss rate impact coefficient and the packet loss rate anomaly assessment value, the traffic packet loss anomaly impact coefficient is calculated.
3. The wireless communication network traffic control system under the edge computing gateway according to claim 2, characterized in that: The specific method for obtaining the shaping delay influence coefficient is as follows: Real-time monitoring and recording of the delay time of each data packet before and after being processed by the shaping algorithm; Calculate the delay difference of each data packet according to the delay time before and after processing, and construct a delay difference sequence; Record the variation range of the delay difference sequence within a period of time and calculate the volatility of the delay difference; Summarize the cumulative delay differences of the shaping algorithm over a period of time to obtain the cumulative delay effect; Based on historical data, define the threshold for delay triggering congestion; When the cumulative delay effect exceeds the threshold of delay triggering congestion, the current time period is marked as a potential congestion period; Perform time series segmentation on the historical shaping data to obtain several time series data, and determine whether they are potential congestion periods, as historical congestion events; Mapping time series data with historical congestion events based on random forests to generate a probability model of delay-triggered congestion; According to the probability model of delay triggering congestion and the volatility of the delay difference, the shaping delay impact coefficient is calculated.
4. The wireless communication network traffic control system under the edge computing gateway according to claim 3 is characterized in that: The buffer size of the shaping algorithm and the bandwidth proportion of each thread are maintained and adjusted based on the traffic shaping risk assessment coefficient, specifically: When the traffic shaping risk assessment coefficient is greater than the pre-examination maintenance threshold, a maintenance request is issued, the buffer size is adjusted arithmetic based on the preset iteration step, and the bandwidth proportion of each thread is internally randomly adjusted to a sum of 1.
5. The wireless communication network traffic control system under the edge computing gateway according to claim 4, characterized in that: The adjustment set draws a three-dimensional adjustment curve, specifically: A three-dimensional rectangular coordinate system is established with the bandwidth ratio of each thread and the buffer size as the x-axis and y-axis, and the traffic shaping risk assessment coefficient as the z-axis; A three-dimensional curve graph is drawn according to the traffic shaping risk assessment coefficient and the corresponding bandwidth ratio of each thread and buffer size. The bandwidth ratio of each thread and buffer size at the lowest point in the curve graph are taken as the optimal buffer size and bandwidth ratio of each thread, and the traffic control of the wireless communication network is adjusted.
6. The wireless communication network traffic control system under the edge computing gateway according to claim 5, characterized in that: The specific calculation formula of the volatility of the delay difference is as follows: The specific calculation formula of the cumulative delay effect is as follows: The specific calculation formula of the shaping delay influence coefficient is as follows: In the formula, is the volatility of the delay difference, is the rth delay difference, is the maximum value in the delay difference sequence, is the minimum value in the delay difference sequence, is the expected delay difference, is the delay difference, is the cumulative delay effect of the vth time series data, is the total number of data in the delay difference sequence, is the shaping delay influence coefficient, Probabilistic model for delay-triggered congestion.
7. The wireless communication network traffic control system under the edge computing gateway according to claim 6, characterized in that: The specific calculation formula of the packet loss rate abnormal evaluation value is as follows: The specific calculation formula of the packet loss rate influence coefficient is as follows: The specific calculation formula of the distribution characteristics of the packet loss rate is as follows: The specific calculation formula of the traffic packet loss abnormality impact coefficient is as follows: In the formula, is the packet loss rate abnormal evaluation value, is the actual packet loss rate, is the expected value of packet loss rate, is the packet loss rate variance, is the packet loss rate influence coefficient, is the covariance of packet loss rate and network load, is the standard deviation of the packet loss rate, is the standard deviation of the network load, is the total number of recorded packet loss rate data, is the bandwidth parameter, is the kernel function, is the pth packet loss rate, is the impact coefficient of abnormal traffic packet loss.
8. The wireless communication network traffic control system under the edge computing gateway according to claim 7, characterized in that: The specific calculation formula of the traffic shaping hidden danger assessment coefficient is as follows: In the formula, is the traffic shaping risk assessment coefficient, is the preset proportional coefficient of the traffic packet loss anomaly impact coefficient, The preset proportionality factor for shaping the delay influence factor, is the preset proportional coefficient of the shaping event abnormality coefficient, is the impact coefficient of abnormal traffic packet loss, The preset proportionality factor for shaping the delay influence factor, It is the preset proportional coefficient of the abnormal coefficient of the shaping event.