A network congestion control method and device based on fuzzy rules

Abstract

The application discloses a network congestion control method and device based on fuzzy rules, and the method comprises the following steps: setting corresponding fuzzy rules according to network types, wherein the network types include a data center network and a wide area network; obtaining network parameters and performing fuzzy processing, fuzzy reasoning and defuzzification on the network parameters to obtain an adjustment factor ΔF, wherein the network parameters include a bandwidth utilization rate, an ECN marking rate and a queue length; obtaining the size of a current congestion window and adjusting the size of the congestion window in real time according to the adjustment factor ΔF; the steps of fuzzy processing and fuzzy reasoning further comprise detecting the current network type, and selecting corresponding fuzzy rules and dynamically adjusting the weight coefficients corresponding to each network parameter in the process of fuzzy reasoning according to the detected network type. The network congestion control method based on fuzzy rules can adjust the network congestion window in real time, realizes the intelligentization and automation of network congestion control, and improves the network management efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer technology, and in particular to a network congestion control method and device based on fuzzy rules. BACKGROUND

[0002] With the rapid development of Internet technology, the number of network service types and services is increasing, and new requirements for real-time data transmission are being put forward. In this case, the performance of the congestion control algorithm, as an integral part of network transmission, directly affects the overall efficiency of the network and user experience.

[0003] Traditional network congestion control algorithms mainly rely on the actual transmission rate and the maximum bandwidth of the port to calculate the bandwidth utilization, and determine the congestion window value through the packet loss rate. This method has some obvious defects, such as: complex calculation and the need for manual adjustment of control parameters, and slow response in dynamic network environments, making it difficult to adapt to network changes, especially in complex environments such as wireless networks and satellite communications, unable to adjust according to different network environments, poor universality and adaptability of the algorithm. In addition, the traditional congestion control algorithm is difficult to accurately distinguish the reasons for the change in delay and the decrease in link throughput and bandwidth utilization, and cannot make reasonable and timely adjustments.

[0004] The above problems make it difficult for traditional congestion control algorithms to meet the needs of today's network environment for transmission, and there is an urgent need for a more intelligent and flexible solution.

[0005] The disclosure of the above background art content is only used to assist in understanding the inventive concept and technical solutions of the present application, and it does not necessarily belong to the prior art of the present application, nor does it necessarily provide technical teaching; in the absence of explicit evidence that the above content has been disclosed before the filing date of the present application, the above background art should not be used to evaluate the novelty and inventiveness of the present application. SUMMARY

[0006] The purpose of the present application is to provide a network congestion control method and device based on fuzzy rules, which can adjust the size of the congestion window in real time according to the network state to realize its flexible adjustment in different network environments, thereby improving the timeliness and accuracy of network congestion control.

[0007] To achieve the above purpose, the technical solution adopted by the present application is as follows:

[0008] A network congestion control method based on fuzzy rules, comprising the following steps:

[0009] According to the network type, set the corresponding fuzzy rule, the network type includes data center network and wide area network;

[0010] Obtain network parameters and perform fuzzification, fuzzy reasoning and de-fuzzification on the network parameters to obtain an adjustment factor ΔF, wherein the network parameters include bandwidth utilization, ECN marking rate and queue length;

[0011] Obtain the size of the current congestion window and adjust the size of the congestion window in real time according to the adjustment factor ΔF;

[0012] The step of fuzzification and the step of fuzzy reasoning further comprise a step of detecting the current network type, wherein corresponding fuzzy rules are selected and the weight coefficients corresponding to each network parameter are dynamically adjusted during the process of fuzzy reasoning according to the detected network type, and when the network type is a data center network, the weight coefficient of the ECN (Explicit Congestion Notification) marking rate is greater than the weight coefficient of the bandwidth utilization; when the network type is a wide area network, the weight coefficient of the bandwidth utilization is greater than the weight coefficient of the ECN marking rate.

[0013] The present application introduces fuzzy rules and presets differentiated fuzzy rules according to the characteristics of different network types, dynamically allocates weight coefficients in combination with network types, adopts a fuzzy reasoning mechanism to adapt to the nonlinear characteristics of network states, and further uses the adjustment factor ΔF obtained from the fuzzy rules to calculate the size of a new congestion window, which is beneficial to overcome the hysteresis of traditional linear algorithms to network mutations, effectively solves the problem of slow response of traditional methods in dynamic network environments, and improves the timeliness and accuracy of network congestion control. At the same time, the control method of the present application forms an integrated control chain of "perception-reasoning-decision-execution" through fuzzy logic and network state feedback closed loop, which can significantly improve the congestion response efficiency in complex network environments, realizes flexible adjustment in different network environments, effectively solves the poor applicability of traditional methods in data centers and wide area networks, and improves the universality and adaptability of the algorithm.

[0014] Further, any of the technical solutions or a combination of the technical solutions described above, the method for obtaining the size of the current congestion window and adjusting the size of the congestion window in real time according to the adjustment factor ΔF is:

[0015] The size of the updated congestion window required for real-time adjustment is calculated according to the following formula,

[0016] W new =W current +β×ΔF×W max

[0017] Wherein, W current is the size of the obtained current congestion window, W max is the maximum window threshold, β is a preset anti-oscillation coefficient, the value range of β is 30%~50%, ΔF is the adjustment factor, the value range of ΔF is -1~1; Wnew to update the size of the congestion window;

[0018] According to the calculation result, the current congestion window W current is adjusted in real time to update the congestion window W new to control network congestion.

[0019] Specifically, W max is determined by the product of network bandwidth and RTT, i.e., W max = Bandwidth (network bandwidth) x RTT (physical link limit). The purpose of setting the anti-oscillation coefficient β is to control the adjustment range, so as to prevent window from oscillating violently and realize smooth scaling of the window, and balance throughput and stability. The value range of the adjustment factor ΔF is -1~1. When it is positive, it indicates that the current network is idle, and the window needs to be expanded. When it is negative, it indicates that there is a risk of network congestion, and the window needs to be reduced.

[0020] Further, any one of the technical solutions or a combination of the technical solutions described above, the value range of W new is greater than or equal to 1 and less than or equal to W max . Wherein, the minimum value of W new is set to 1, in order to ensure that the sending end will not stop sending packets completely in the most extreme congestion situation, to ensure the activity of the protocol and prevent deadlock, which is the basic safety bottom line of congestion control.

[0021] Further, any one of the technical solutions or a combination of the technical solutions described above, the bandwidth utilization rate, the ECN marking rate and the queue length are input parameters of the fuzzy rule, and the fuzzy output of the fuzzy rule is a fuzzy adjustment level, which includes a first level, a second level, a third level, a fourth level and a fifth level.

[0022] Further, any one of the technical solutions or a combination of the technical solutions described above, the method for setting the fuzzy rule is:

[0023] Each of the network parameters of the bandwidth utilization rate, the ECN marking rate and the queue length is set to low, lower, medium, higher and high respectively, and all combinations of the levels of the bandwidth utilization rate, the levels of the ECN marking rate and the levels of the queue length and the fuzzy adjustment level corresponding to each combination are listed.

[0024] Further, any one of the above technical solutions or a combination of multiple technical solutions, the value range corresponding to the low level of bandwidth utilization is 0-0.3, the value range corresponding to the lower level of bandwidth utilization is 0.2-0.45, the value range corresponding to the medium level of bandwidth utilization is 0.35-0.65, the value range corresponding to the higher level of bandwidth utilization is 0.55-0.8, and the value range corresponding to the high level of bandwidth utilization is 0.7-1.

[0025] Further, any one of the above technical solutions or a combination of multiple technical solutions, the value range corresponding to the low level of ECN marking rate is 0-1, the value range corresponding to the lower level of ECN marking rate is 0.5-3, the value range corresponding to the medium level of ECN marking rate is 2-6, the value range corresponding to the higher level of ECN marking rate is 5-15, and the value range corresponding to the high level of ECN marking rate is 10-100.

[0026] Further, any one of the above technical solutions or a combination of multiple technical solutions, the value range corresponding to the low level of queue length is 0-5, the value range corresponding to the lower level of queue length is 3-15, the value range corresponding to the medium level of queue length is 10-30, the value range corresponding to the higher level of queue length is 20-60, and the value range corresponding to the high level of queue length is 50-70.

[0027] Further, any one of the above technical solutions or a combination of multiple technical solutions, when the network type is a data center network, the corresponding fuzzy rule is a first fuzzy rule; when the network type is a wide area network, the corresponding fuzzy rule is a second fuzzy rule, and the combination of the same level of bandwidth utilization, ECN marking rate, and queue length in the first fuzzy rule and the second fuzzy rule corresponds to different or same fuzzy adjustment levels. Since the characteristics of the data center network are high bandwidth and low delay sensitivity, and the characteristics of the wide area network are high fluctuation and long delay, by pre-setting differentiated fuzzy rules, the bandwidth utilization efficiency and congestion prevention demand are effectively balanced.

[0028] Further, any one of the above technical solutions or a combination of multiple technical solutions, the first fuzzy rule and the second fuzzy rule each include the same number of sub-rules, and the number of all combinations of the level of bandwidth utilization, the level of ECN marking rate, and the level of queue length is equal to the number of sub-rules in the fuzzy rule.

[0029] Further, any one of the above technical solutions or a combination of the above technical solutions, the method for detecting the current network type is:

[0030] Obtaining the round-trip delay, delay jitter, bandwidth, queue change rate and flow duration of the current network as input parameters, and setting a first score variable and a second score variable, wherein the first score variable is used for feature matching score of the data center network, and the second score variable is used for feature matching score of the wide area network;

[0031] When the round-trip delay is less than 1ms and the delay jitter is less than 0.2ms, the weight of the first score variable is increased by 0.4, otherwise the weight of the second score variable is increased by 0.4;

[0032] When the bandwidth is greater than or equal to 10Gbps and the queue change rate is greater than 1000 packets / ms, the weight of the first score variable is increased by 0.3, otherwise the weight of the second score variable is increased by 0.3;

[0033] When the flow duration is less than 0.1s, the weight of the first score variable is increased by 0.3, otherwise the weight of the second score variable is increased by 0.3;

[0034] Finally, according to the size relationship between the first score variable and the second score variable, the network type is judged, when the first score variable is greater than the second score variable, it is identified as a data center network, and when the first score variable is less than the second score variable, it is identified as a wide area network.

[0035] Further, any one of the above technical solutions or a combination of the above technical solutions, the method for fuzzy processing the network parameters is:

[0036] According to the accurate value of each network parameter obtained and the fuzzy rule, each network parameter is corresponded to the grade to which it belongs, wherein each network parameter corresponds to at least one grade.

[0037] Further, any one of the above technical solutions or a combination of the above technical solutions, the method for fuzzy reasoning the network parameters is:

[0038] According to the current network type detected, the original weight coefficients of the bandwidth utilization rate, the ECN marking rate and the queue length are preset, and after the normalization of each original weight coefficient, the corresponding normalized weight coefficients are obtained;

[0039] According to the membership function, the membership degrees corresponding to the bandwidth utilization rate, the ECN marking rate and the queue length are calculated respectively;

[0040] According to the normalized weight coefficients and the respective membership degrees corresponding to the bandwidth utilization rate, the ECN marking rate and the queue length, the activation degree is calculated.

[0041] Further, any one of the technical solutions or the combination of the multiple technical solutions, the normalized weight coefficient is calculated by the following formula:

[0042]

[0043] Wherein, w are the original weight coefficients of bandwidth utilization rate, ECN marking rate and queue length respectively, w i represents the normalized weight coefficient of the i th network parameter, i = 1, 2, 3.

[0044] Further, any one of the technical solutions or the combination of the multiple technical solutions, the membership degree is calculated by the following formula:

[0045]

[0046] Wherein, x represents the accurate value of the obtained network parameter, a is the width parameter, b is the steepness parameter, c is the membership function center, and μ is the membership degree corresponding to the network parameter. Specifically, the width parameter a can be used to distinguish the overlapping width of the fuzzy set, and the steepness parameter b is in the range of 2-3.

[0047] Further, any one of the technical solutions or the combination of the multiple technical solutions, when the network type is a data center network, the is 0.2-0.3, the is 0.4-0.6, the is 0.2-0.3; when the network type is a wide area network, the is 0.4-0.6, the is 0.2-0.3, the is 0.2-0.3. In the application, in the process of fuzzy reasoning, the original weight coefficients corresponding to the network parameters are dynamically allocated according to different network types. Specifically, when the network type is a data center network, the original weight coefficient of the ECN marking rate is greater than the original weight coefficient of the bandwidth utilization rate; when the network type is a wide area network, the original weight coefficient of the bandwidth utilization rate is greater than the original weight coefficient of the ECN marking rate, so as to play a weighting role in the fuzzy rule, so that the fuzzy rule can better fit the application scenario.

[0048] Further, any one of the technical solutions or the combination of the multiple technical solutions, the activation degree is calculated by the following formula:

[0049] α=w1×μ1+w2×μ2+w3×μ3

[0050] Wherein, a is the activation degree, w1 is the normalized weight coefficient of bandwidth utilization rate, and μ1 is the membership degree of bandwidth utilization rate; w2 is the normalized weight coefficient of ECN marking rate, and μ2 is the membership degree of ECN marking rate; w3 is the normalized weight coefficient of queue length, and μ3 is the membership degree of queue length.

[0051] Further, any one of the preceding technical solutions or a combination of the preceding technical solutions, the method for de-fuzzifying network parameters comprises the following steps.

[0052] According to the detected current network type, a corresponding fuzzy rule is selected, in the corresponding fuzzy rule, a corresponding sub-rule is confirmed according to the grade of each network parameter obtained in the fuzzy processing process, and the fuzzy adjustment grade in the current sub-rule is converted into a corresponding representative value, and then the adjustment factor ΔF is calculated according to the following formula:

[0053]

[0054] Wherein, m represents the total number of activated sub-rules in the corresponding fuzzy rule under the grade combination of each network parameter input currently, a j is the activation degree of the jth sub-rule in all activated sub-rules (m sub-rules), F j is the representative value of the fuzzy adjustment grade corresponding to the jth sub-rule in all activated sub-rules (m sub-rules).

[0055] Further, any one of the preceding technical solutions or a combination of the preceding technical solutions, when the fuzzy adjustment grade is the first grade, the value of Fj corresponding thereto is -1, when the fuzzy adjustment grade is the second grade, the value of Fj corresponding thereto is -0.5, when the fuzzy adjustment grade is the third grade, the value of Fj corresponding thereto is 0, when the fuzzy adjustment grade is the fourth grade, the value of Fj corresponding thereto is 0.5, and when the fuzzy adjustment grade is the fifth grade, the value of Fj corresponding thereto is 1.

[0056] Further, any one of the preceding technical solutions or a combination of the preceding technical solutions, further comprising the following steps:

[0057] According to the size of the updated congestion window, the number of data packets simultaneously sent by the sending end is determined, and the data packets are put into the sending queue for sending, and the network state is monitored, and the foregoing steps are performed every preset time interval to obtain the size of the new congestion window, so that the adjustment is constantly and real-timely updated. In some embodiments of the present application, the preset time interval is 50-100 ms. In addition, according to the size of the updated congestion window, the number of data packets simultaneously sent by the sending end is determined, which can adjust the buffer pressure of the intermediate device (such as a router) to avoid continuous packet loss and delay deterioration.

[0058] According to another aspect of the present application, there is provided a network congestion control device for controlling network congestion by using the network congestion control method as described above, the network congestion control device comprising a fuzzy decision module and a data analysis module, the fuzzy decision module being configured to detect a current network type and determine a fuzzy adjustment level outputted in a fuzzy rule according to obtained network parameters, and the data analysis module being configured to receive and analyze data packets sent by a transport layer.

[0059] The technical solution provided by the present application has the following beneficial effects:

[0060] a. The present application realizes the intelligentization and automation of network congestion control by fusing multi-dimensional network parameters, combining the weighted moving average processing of historical data, constructing a precise network state quantization model, and processing the network state by fuzzy rules, avoiding the complexity and inconvenience of manual parameter adjustment in traditional methods, and greatly improving the network management efficiency;

[0061] b. The present application presets different fuzzy rules corresponding to data center networks and wide area networks according to their different characteristics, realizes flexible adjustment in different network environments by the multiple judgment mechanism of fuzzy rules and the dynamic allocation of weight coefficients according to network types, solves the problems of slow response in dynamic network environment and poor applicability in data center networks and wide area networks in traditional methods, and improves the universality and adaptability of the algorithm;

[0062] c. The present application fuzzes network parameters into different levels through nonlinear fuzzy reasoning, and finally outputs an adjustment factor ΔF for real-time adjustment of the size of the congestion window through fuzzy reasoning, which is beneficial to overcome the hysteresis of traditional linear algorithms to network mutations, can realize precise elastic adjustment of the congestion window, and significantly improve the congestion response efficiency in complex network environment. BRIEF DESCRIPTION OF DRAWINGS

[0063] In order to more clearly illustrate the technical solutions in the embodiments or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments described in the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0064] Figure 1 The flowchart of the congestion control method based on fuzzy rules is provided for an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0065] In the following, the technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application in order to make the technical personnel in the technical field better understand the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by the ordinary technical personnel in the technical field without creative labor should belong to the protection scope of the present application.

[0066] It should be noted that the terms "first", "second" and the like in the description and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, device, product or apparatus that includes a series of steps or units need not be limited to those steps or units clearly listed, but can include other steps or units that are not clearly listed or inherent to such processes, methods, products or apparatus.

[0067] Referring to Figure 1 In an embodiment of the present application, a network congestion control method based on fuzzy rules is provided, comprising the following steps:

[0068] Step 1: Set a first fuzzy rule corresponding to a data center network and a second fuzzy rule corresponding to a wide area network, respectively, wherein the first fuzzy rule and the second fuzzy rule each include a same number of multiple sub-rules; for the first fuzzy rule and the second fuzzy rule, bandwidth utilization rate, ECN marking rate and queue length are input parameters, and the fuzzy output of each fuzzy rule is a fuzzy adjustment level.

[0069] Specifically, each of the network parameters of bandwidth utilization rate, queue length and ECN marking rate is respectively set to five levels of low, lower, medium, higher and high, the fuzzy adjustment levels include a first level, a second level, a third level, a fourth level and a fifth level, all combinations of the levels of bandwidth utilization rate, the levels of queue length and the levels of ECN marking rate and the corresponding fuzzy adjustment levels of each combination are listed, the number of all combinations of the levels of bandwidth utilization rate, the levels of queue length and the levels of ECN marking rate is equal to the number of sub-rules in the fuzzy rule, the first fuzzy rule in the application is specifically shown in Table 1, the second fuzzy rule is specifically shown in Table 2, and the two fuzzy rules each include 125 sub-rules. As can be seen from Tables 1 and 2, the fuzzy adjustment levels corresponding to the same combination of the levels of bandwidth utilization rate, the levels of queue length and the levels of ECN marking rate in the first fuzzy rule and the second fuzzy rule may be different or the same, because the first fuzzy rule and the second fuzzy rule are respectively set based on the data center network and the wide area network, and the two network types have different characteristics, the characteristics of the data center network are high bandwidth and low delay sensitivity, and the characteristics of the wide area network are high fluctuation and long delay, by presetting the first fuzzy rule and the second fuzzy rule with different characteristics in advance, the bandwidth utilization efficiency and the congestion prevention demand can be effectively balanced.

[0070] For the first fuzzy rule and the second fuzzy rule, the value range of the same level of bandwidth utilization rate is the same: when the level of bandwidth utilization rate is low, the corresponding value range is 0-0.3, when the level of bandwidth utilization rate is lower, the corresponding value range is 0.2-0.45, when the level of bandwidth utilization rate is medium, the corresponding value range is 0.35-0.65, when the level of bandwidth utilization rate is higher, the corresponding value range is 0.55-0.8, and when the level of bandwidth utilization rate is high, the corresponding value range is 0.7-1.

[0071] For the first fuzzy rule and the second fuzzy rule, the value range of the same level of ECN marking rate is the same: when the level of ECN marking rate is low, the corresponding value range is 0-1, when the level of ECN marking rate is lower, the corresponding value range is 0.5-3, when the level of ECN marking rate is medium, the corresponding value range is 2-6, when the level of ECN marking rate is higher, the corresponding value range is 5-15, and when the level of ECN marking rate is high, the corresponding value range is 10-100.

[0072] For the first fuzzy rule and the second fuzzy rule, the value ranges of the same level of queue length are the same: when the level of queue length is low, the corresponding value range is 0-5; when the level of queue length is lower, the corresponding value range is 3-15; when the level of queue length is medium, the corresponding value range is 10-30; when the level of queue length is higher, the corresponding value range is 20-60; and when the level of queue length is high, the corresponding value range is 50-70.

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079] Step 2: Obtain network parameters, and perform fuzzification, fuzzy inference and defuzzification on the network parameters to obtain an adjustment factor AF.

[0080] Step 2.1: Receive a data packet sent by a network transmission layer, and parse the data packet to obtain network parameters of the data packet, the network parameters including a bandwidth utilization rate, an ECN marking rate and a queue length.

[0081] Step 2.2: Fuzzify the network parameters.

[0082] According to the obtained accurate values of each network parameter and the fuzzy rules, each network parameter is corresponded to the level to which it belongs, wherein each network parameter corresponds to at least one level. Since there is an overlapping part in the value ranges of different levels of the same network parameter, when the accurate value of a certain network parameter falls within the value ranges of two levels, the sub-rules of the network parameter corresponding to the two levels need to be listed respectively and applied to the subsequent calculation process. Similarly, whether one, two or three of the three obtained network parameters can correspond to different levels, all the sub-rules need to be listed respectively.

[0083] Step 2.3: Detect the current network type.

[0084] The round-trip delay, delay jitter, bandwidth, queue change rate and flow duration of the current network are obtained as input parameters, and a first score variable and a second score variable are set, wherein the first score variable is used for feature matching score of the data center network, and the second score variable is used for feature matching score of the wide area network.

[0085] When the round-trip delay is less than 1ms and the delay jitter is less than 0.2ms, the weight of the first score variable is increased by 0.4, otherwise the weight of the second score variable is increased by 0.4; when the bandwidth is greater than or equal to 10Gbps and the queue change rate is greater than 1000 packets / ms, the weight of the first score variable is increased by 0.3, otherwise the weight of the second score variable is increased by 0.3; when the flow duration is less than 0.1s, the weight of the first score variable is increased by 0.3, otherwise the weight of the second score variable is increased by 0.3.

[0086] Finally, the network type is determined according to the size relationship between the first score variable and the second score variable, and when the first score variable is greater than the second score variable, it is identified as a data center network, and when the first score variable is less than the second score variable, it is identified as a wide area network, so as to determine the current network type, so as to select the corresponding fuzzy rule in subsequent calculation.

[0087] Step 2.4: fuzzy reasoning on network parameters.

[0088] According to the detected current network type, the original weight coefficients of the bandwidth utilization rate, the ECN marking rate and the queue length are preset, and the original weight coefficients are normalized according to the following formula to obtain the corresponding normalized weight coefficients:

[0089]

[0090] wherein, are the original weight coefficients of the bandwidth utilization rate, the ECN marking rate and the queue length, w i represents the normalized weight coefficient of the i th network parameter, i = 1, 2, 3.

[0091] When the network type is a data center network, the weight coefficient of the ECN marking rate is greater than the weight coefficient of the bandwidth utilization rate due to the sensitivity of the queue, and the weight coefficient of the bandwidth utilization rate is greater than the weight coefficient of the queue length due to the sensitivity of the queue, and is 0.2~0.3, is 0.4~0.6, is 0.2~0.3; when the network type is a wide area network, the weight coefficient of the bandwidth utilization rate is greater than the weight coefficient of the ECN marking rate due to the need for long-term fairness, and the weight coefficient of the ECN marking rate is greater than the weight coefficient of the queue length due to the need for long-term fairness, and is 0.4~0.6, is 0.2~0.3, The original weight coefficient corresponding to each network parameter is dynamically allocated according to different network types, so as to play a weighting role in the fuzzy rule, and the fuzzy rule can better fit the application scenario.

[0092] According to the membership function, the membership degrees corresponding to the bandwidth utilization rate, the ECN marking rate and the queue length are calculated respectively according to the following formula:

[0093]

[0094] Wherein, x represents the accurate value of the obtained network parameter, a is a width parameter, which can be used to distinguish the overlapping width of the fuzzy set; b is a steepness parameter, and the value range of b is 2-3; c is a membership function center, and μ is the membership degree corresponding to the network parameter.

[0095] The activation is calculated according to the following formula according to the normalized weight coefficient corresponding to the bandwidth utilization rate, the ECN marking rate and the queue length and the respective corresponding membership degrees:

[0096] α = w1 x μ1 + w2 x μ2 + w3 x μ3

[0097] Wherein, α is the activation, w1 is the normalized weight coefficient of the bandwidth utilization rate, μ1 is the membership degree of the bandwidth utilization rate; w2 is the normalized weight coefficient of the ECN marking rate, μ2 is the membership degree of the ECN marking rate; w3 is the normalized weight coefficient of the queue length, and μ3 is the membership degree of the queue length.

[0098] Step 2.5: Defuzzification of network parameters.

[0099] According to the detected current network type, the first fuzzy rule or the second fuzzy rule is selected, the corresponding subrule is confirmed according to the grade of each network parameter obtained in the fuzzy processing process in the corresponding fuzzy rule, and the fuzzy adjustment grade in the current subrule is converted into the corresponding representative value: when the fuzzy adjustment grade is the first grade, the value of the corresponding representative value is-1, when the fuzzy adjustment grade is the second grade, the value of the corresponding representative value is-0.5, when the fuzzy adjustment grade is the third grade, the value of the corresponding representative value is 0, when the fuzzy adjustment grade is the fourth grade, the value of the corresponding representative value is 0.5, and when the fuzzy adjustment grade is the fifth grade, the value of the corresponding representative value is 1; and the adjustment factor ΔF is calculated according to the following formula:

[0100]

[0101] Wherein, m represents the total number of subrules activated under the grade combination of each network parameter in the corresponding fuzzy rule, α jF is the activation degree of the jth sub-rule in the m activated sub-rules j F is the representative value of the fuzzy adjustment level corresponding to the jth sub-rule in the m activated sub-rules.

[0102] Step 3: Obtain the size of the current congestion window W current , calculate the size of the updated congestion window W new required for real-time adjustment, and then real-time adjust the current congestion window W current to the updated congestion window W new according to the calculation result, W new .

[0103] W new =W current +β×ΔF×W max

[0104] Wherein, W current is the size of the obtained current congestion window, W max is the maximum window threshold, which is determined by the product of network bandwidth and delay, that is, W max =Bandwidth×RTT; β is a preset anti-oscillation coefficient, the value range of β is 30%-50%, the introduction of anti-oscillation coefficient β in the calculation formula can control the adjustment amplitude, prevent window from violent oscillation, and realize window smooth scaling; ΔF is an adjustment factor, the value range of ΔF is-1-1; W new is the size of the updated congestion window, and the value range of W new is greater than or equal to 1 and less than or equal to W max .

[0105] Step 4: Determine the number of data packets sent by the sending end at the same time according to the size of the updated congestion window, and then put the data packets into the sending queue for sending, and monitor the network state, execute steps 2 and 3 every preset time interval to obtain the size of the new congestion window again, so as to continuously and real-time update and adjust. Wherein, the preset time interval is 50-100ms.

[0106] Another embodiment of the application also provides a device for network congestion control using the above network congestion control method, which realizes precise and elastic adjustment of the congestion window by setting a fuzzy decision module and a data analysis module, real-time analyzing multi-dimensional network parameters such as bandwidth utilization rate, ECN marking rate and queue length, and dynamically allocating weight coefficients combined with network types.

[0107] Specifically, the fuzzy decision module in the network congestion control device is configured to detect the current network type and determine the fuzzy adjustment level output in the fuzzy rules based on the obtained network parameters. The fuzzy decision module includes a fuzzification processor, a knowledge base, an inference engine, and a defuzzifier. The fuzzification processor converts input network parameters (bandwidth utilization, ECN marking rate, and queue length) into linguistic variables (i.e., the corresponding levels of the network parameters: low, lower, medium, higher, and high). It also calculates the precise values ​​of the input network parameters based on a preset membership function to obtain the corresponding membership. The knowledge base stores preset first and second fuzzy rules. The inference engine matches the input variables (membership values ​​and normalized weight coefficients) with the fuzzy rules, calculates the activation degree of each sub-rule, and determines the output fuzzy set. The defuzzifier converts the fuzzy output (fuzzy adjustment level) into a clear numerical output (adjustment factor ΔF).

[0108] The data analysis module is configured to receive and parse data packets sent by the transport layer. It includes an ECN marking counter, a total ACK counter, an ECN marking rate calculator, a port utilization collector, an ECN field parser, a network layer round-trip delay and packet loss rate detector, and a queue smoothing calculator. The ECN marking counter continuously records the number of ECN-marked packets received per unit time. It parses the ECN field in the IP packet header to determine whether a packet is congestion-marked and accumulates the statistics within each time window. The total ACK (acknowledgement packet) counter records the total number of ACKs received during the current observation period, providing an accurate denominator for subsequent calculations of the ECN marking rate, reflecting the proportion of marked congestion in the network. The ECN marking rate calculator calculates the ECN marking rate per unit time (the number of ECN-marked packets divided by the total number of ACK packets). The port utilization collector obtains real-time bandwidth usage on ports of network devices (such as switches and routers) to reflect current link utilization. The ECN field parser is used to analyze the ECN field status in data packets in real time. By checking the field bits on each inbound or outbound packet, it records congestion trends and assists in determining path health and whether explicit congestion feedback exists on the current path. The network layer round-trip delay and packet loss rate acquirer estimates current delay and packet loss using methods such as the TCP (Transmission Control Protocol) timestamp option, ICMP (Control Message Protocol) echo requests, link-layer statistics, or stack measurements. This provides a basis for dynamic window adjustment, thereby monitoring and acquiring the round-trip delay and packet loss rate of the path. The queue smoothing calculator uses historical records to perform weighted smoothing on the queue length (using a weighted moving average to incorporate the current queue length) to mitigate misjudgments caused by transient fluctuations.

[0109] The fuzzy rule-based congestion control method of the present invention introduces fuzzy rules and pre-defines differentiated fuzzy rules based on the characteristics of different network types. It dynamically assigns weight coefficients based on network types and uses a fuzzy inference mechanism to adapt to the nonlinear characteristics of network states. By further using the adjustment factor ΔF derived from the fuzzy rules to calculate the new congestion window size, this method overcomes the lag inherent in traditional linear algorithms in responding to sudden network changes, effectively addressing the sluggish response of traditional methods in dynamic network environments and improving the timeliness and accuracy of network congestion control. Furthermore, by forming a closed loop with fuzzy logic and network state feedback, it significantly improves congestion response efficiency in complex network environments, enabling flexible adjustment in diverse network environments and enhancing the algorithm's versatility and adaptability.

[0110] Due to the advanced nature of this technical solution, the technology of the present invention can be widely used in scenarios such as cloud computing data centers, real-time video transmission, and large-scale Internet of Things. In particular, in 5G edge computing and SDN network architectures, it can significantly reduce network latency fluctuations and improve link utilization in high-concurrency scenarios. At the same time, through the dynamic window adjustment mechanism, it avoids the "sawtooth" throughput fluctuations of the traditional TCP protocol, providing a low-latency, high-throughput, and highly adaptive core congestion control solution for the new generation of intelligent networks.

[0111] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0112] The above is only a specific implementation method of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method of network congestion control based on fuzzy rules, characterized by, The method comprises the following steps: According to the network type, the corresponding fuzzy rule is set, and the network type comprises a data center network and a wide area network; Network parameters are obtained and fuzzy processing, fuzzy reasoning and de-fuzzification are performed on the network parameters to obtain an adjustment factor ΔF, and the network parameters comprise a bandwidth utilization rate, an ECN marking rate and a queue length; The size of the current congestion window is obtained, and the size of the congestion window is adjusted in real time according to the adjustment factor ΔF, and the method is: The size of the updated congestion window required for real-time adjustment is calculated according to the following formula, W new = W current + β · ΔF · W max Wherein, W current is the size of the current congestion window obtained, W max is the maximum window threshold, β is a preset anti-oscillation coefficient, the value range of β is 30%~50%, ΔF is an adjustment factor, the value range of ΔF is -1~1; W new is the size of the updated congestion window; and the current congestion window W current is adjusted in real time to the updated congestion window W new to control network congestion. The step of fuzzy processing and the step of fuzzy reasoning further comprise the step of detecting the current network type, and in the process of fuzzy reasoning, the corresponding fuzzy rule is selected according to the detected network type and the weight coefficient corresponding to each network parameter is dynamically adjusted, and when the network type is a data center network, the weight coefficient of the ECN marking rate is greater than the weight coefficient of the bandwidth utilization rate; when the network type is a wide area network, the weight coefficient of the bandwidth utilization rate is greater than the weight coefficient of the ECN marking rate.

2. The network congestion control method of claim 1, wherein, The W new value range is greater than or equal to 1 and less than or equal to W max .

3. The network congestion control method of claim 1, wherein, The bandwidth utilization rate, the ECN marking rate and the queue length are all input parameters of the fuzzy rule, the fuzzy output of the fuzzy rule is a fuzzy adjustment level, and the fuzzy adjustment level comprises a first level, a second level, a third level, a fourth level and a fifth level.

4. The network congestion control method of claim 3, wherein, The method for setting the fuzzy rule is: Each network parameter in the bandwidth utilization rate, the ECN marking rate and the queue length is respectively set to five levels of low, lower, medium, higher and high, all combinations of the levels of the bandwidth utilization rate, the levels of the ECN marking rate and the levels of the queue length are listed, and the fuzzy adjustment level corresponding to each combination is listed.

5. The network congestion control method of claim 4, wherein, When the level of the bandwidth utilization rate is low, the corresponding value range is 0~0.3, when the level of the bandwidth utilization rate is lower, the corresponding value range is 0.2~0.45, when the level of the bandwidth utilization rate is medium, the corresponding value range is 0.35~0.65, when the level of the bandwidth utilization rate is higher, the corresponding value range is 0.55~0.8, and when the level of the bandwidth utilization rate is high, the corresponding value range is 0.7~1.

6. The network congestion control method of claim 4, wherein, When the level of the ECN marking rate is low, the corresponding value range is 0~1, when the level of the ECN marking rate is lower, the corresponding value range is 0.5~3, when the level of the ECN marking rate is medium, the corresponding value range is 2~6, when the level of the ECN marking rate is higher, the corresponding value range is 5~15, and when the level of the ECN marking rate is high, the corresponding value range is 10~100.

7. The network congestion control method of claim 4, wherein, When the level of the queue length is low, the corresponding value range is 0~5, when the level of the queue length is lower, the corresponding value range is 3~15, when the level of the queue length is medium, the corresponding value range is 10~30, when the level of the queue length is higher, the corresponding value range is 20~60, and when the level of the queue length is high, the corresponding value range is 50~70.

8. The network congestion control method of claim 4, wherein, When the network type is a data center network, the corresponding fuzzy rule is a first fuzzy rule; when the network type is a wide area network, the corresponding fuzzy rule is a second fuzzy rule, and the combination of the same level of bandwidth utilization rate, the level of ECN marking rate and the level of queue length in the first fuzzy rule and the second fuzzy rule corresponds to different or same fuzzy adjustment levels.

9. The network congestion control method of claim 8, wherein, The first fuzzy rule and the second fuzzy rule each include the same number of sub-rules, and the number of all combinations of the level of bandwidth utilization rate, the level of ECN marking rate and the level of queue length is equal to the number of sub-rules in the fuzzy rule.

10. The network congestion control method of claim 1, wherein, The method for detecting the current network type is: Obtaining the round-trip delay, delay jitter, bandwidth, queue change rate and flow duration of the current network as input parameters, and setting a first score variable and a second score variable, wherein the first score variable is used for feature matching score of the data center network, and the second score variable is used for feature matching score of the wide area network; When the round-trip delay is less than 1ms and the delay jitter is less than 0.2ms, the weighting of the first score variable is increased by 0.4, otherwise the weighting of the second score variable is increased by 0.4; When the bandwidth is greater than or equal to 10Gbps and the queue change rate is greater than 1000 packets / ms, the weighting of the first score variable is increased by 0.3, otherwise the weighting of the second score variable is increased by 0.3; When the flow duration is less than 0.1s, the weighting of the first score variable is increased by 0.3, otherwise the weighting of the second score variable is increased by 0.3; Finally, the network type is determined according to the size relationship between the first score variable and the second score variable, and when the first score variable is greater than the second score variable, the network type is identified as a data center network, and when the first score variable is less than the second score variable, the network type is identified as a wide area network.

11. The network congestion control method of claim 4, wherein, The method for fuzzy processing of network parameters is: According to the obtained accurate value of each network parameter and the fuzzy rule, each network parameter is corresponded to the level to which it belongs, wherein each network parameter corresponds to at least one level.

12. The network congestion control method of claim 11, wherein, The method for fuzzy reasoning of network parameters is: According to the preset original weight coefficients of bandwidth utilization rate, ECN marking rate and queue length according to the detected current network type, and after normalizing each original weight coefficient, corresponding normalized weight coefficients are obtained respectively; According to the membership function, the membership degrees corresponding to the bandwidth utilization rate, the ECN marking rate and the queue length are calculated respectively; The activation degree is calculated according to the normalized weight coefficients and the corresponding membership degrees of the bandwidth utilization rate, the ECN marking rate and the queue length.

13. The network congestion control method of claim 12, wherein, The normalized weight coefficient is calculated by the following formula: ； wherein, , , are the original weight coefficients of bandwidth utilization, ECN marking rate and queue length, respectively, w i denotes the normalized weight coefficient of the i-th network parameter, i = 1, 2, 3.

14. The method of network congestion control according to claim 13, wherein, The membership degree is calculated by the following formula: ； Wherein, x represents the accurate value of the obtained network parameter, a is the width parameter, b is the steepness parameter, c is the membership function center, and μ is the membership degree corresponding to the network parameter.

15. The network congestion control method according to claim 13, characterized in that: When the network type is a data center network, the is 0.2-0.3, the is 0.4-0.6, the is 0.2-0.3; when the network type is a wide area network, the is 0.4-0.6, the is 0.2-0.3, the is 0.2-0.

3.

16. The network congestion control method of claim 14, wherein, The activation degree is calculated by the following formula: α=w1·μ1+ w2·μ2+w3·μ3 Wherein, a is the activation degree, w1 is the normalized weight coefficient of bandwidth utilization, μ1 is the membership degree of bandwidth utilization; w2 is the normalized weight coefficient of ECN marking rate, μ2 is the membership degree of ECN marking rate; w3 is the normalized weight coefficient of queue length, μ3 is the membership degree of queue length.

17. The method of network congestion control according to claim 16, wherein, The method for de-fuzzing network parameters is: According to the detected current network type, a corresponding fuzzy rule is selected, in the corresponding fuzzy rule, a corresponding sub-rule is confirmed according to the grade of each network parameter obtained in the fuzzy processing process, and the fuzzy adjustment grade in the current sub-rule is converted into a corresponding representative value, and then an adjustment factor ΔF is calculated according to the following formula: ； wherein m represents the total number of sub-rules activated under the level combination of each network parameter of the current input in the corresponding fuzzy rule, a j is the activation degree of the jth sub-rule in all activated sub-rules, F j is the representative value of the fuzzy adjustment level corresponding to the jth sub-rule in all activated sub-rules.

18. The method of network congestion control according to claim 17, wherein, When the fuzzy adjustment grade is the first grade, the value of Fj corresponding thereto is -1, when the fuzzy adjustment grade is the second grade, the value of Fj corresponding thereto is -0.5, when the fuzzy adjustment grade is the third grade, the value of Fj corresponding thereto is 0, when the fuzzy adjustment grade is the fourth grade, the value of Fj corresponding thereto is 0.5, and when the fuzzy adjustment grade is the fifth grade, the value of Fj corresponding thereto is 1.

19. A network congestion control apparatus, characterized by, The network congestion control device utilizes the network congestion control method according to any one of claims 1-18 to perform network congestion control, and the network congestion control device comprises a fuzzy decision module and a data analysis module, the fuzzy decision module is configured to detect the current network type and determine the fuzzy adjustment grade output in the fuzzy rule according to the obtained network parameters, and the data analysis module is configured to receive the data packet sent by the transmission layer and analyze the data packet.

Images