Flow data processing method and system based on active noise reduction

By calculating the impact of comprehensive traffic data of network security equipment, based on the active noise reduction processing method, the problem of false alarms of network security equipment is solved, accurate identification of network threats and reduced alarm volume, and has a small impact on network performance. .

CN118890199AActive Publication Date: 2024-11-01SHANXI CHINA MOBILE COMM CORP +1
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202411098853.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-12
Publication Date
2024-11-01
Estimated Expiration
2044-08-12

AI Technical Summary

Technical Problem

Existing network security equipment generates a large number of false alarms, resulting in the real network intrusion being ignored. It is difficult for the existing technology to effectively reduce the number of alarms and accurately identify threat behaviors.

Method used

Adopt traffic data processing method based on active noise reduction, obtain alarm information data of network security equipment, calculate the impact of comprehensive traffic data, and perform active noise reduction processing based on the maximum number of attacks and attack half-life, including prohibited traffic data and random blocking. operate.

Benefits of technology

It reduces the alarm volume of network security devices, accurately identify current threat behaviors, reduces false positive rates, and has a small impact on network performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118890199A_ABST
    Figure CN118890199A_ABST
Patent Text Reader

Abstract

The invention provides a traffic data processing method and system based on active noise reduction. The method comprises the following steps: acquiring alarm information data generated by network security equipment at the current moment; according to the alarm information data, determining the maximum number of attack times of network attacks corresponding to each network attack type received by the network security equipment in a preset period and the attack half-life period of the network attacks corresponding to each network attack type, according to the maximum attack times and the attack half-life periods of the network attacks of all the network attack types, the comprehensive flow data influence degree of the network security equipment at the current moment is obtained through calculation; and under the condition that the influence degree of the comprehensive traffic data is determined to be greater than a preset influence degree threshold value, performing corresponding active noise reduction processing on the traffic data of the network security equipment. According to the invention, the alarm quantity of the network security equipment is reduced, and the real threat behavior of the current security equipment is accurately identified.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of network security technology, and in particular to a flow data processing method and system based on active noise reduction. Background Art

[0002] In the current digital environment, network security is of vital importance. Network security devices such as firewalls, intrusion detection systems (IDS), and intrusion prevention systems (IPS) are widely used to monitor and protect networks from various threats. These security devices often generate a large number of alarms, including many false alarms.

[0003] In existing related technologies, such as rule-based detection and signature detection, a high false alarm rate is often generated, which may cause network administrators to ignore real threats and cause the actual network intrusion to be overwhelmed by a large number of false alarms.

[0004] Therefore, there is an urgent need for a flow data processing method and system based on active noise reduction to solve the above problems. Summary of the invention

[0005] In view of the problems existing in the prior art, the present invention provides a flow data processing method and system based on active noise reduction.

[0006] The present invention provides a flow data processing method based on active noise reduction, comprising: Obtain the alarm information data generated by the network security device at the current moment; Determine, based on the alarm information data, the maximum number of network attacks corresponding to each network attack type suffered by the network security device within a preset period, and the attack half-life of each network attack type corresponding to the network attack, and calculate the comprehensive traffic data impact of the network security device at the current moment based on the maximum number of attacks and the attack half-life of network attacks of all network attack types; When it is determined that the impact of the comprehensive traffic data is greater than a preset impact threshold, performing corresponding active noise reduction processing on the traffic data of the network security device; Among them, the maximum number of attacks is obtained based on the number of attacks on the network security device at each historical moment within a preset period; the attack half-life represents the length of time required for the number of attacks to reach half of the corresponding maximum number of attacks when the network security device is attacked by a network within a preset period.

[0007] According to a flow data processing method based on active noise reduction provided by the present invention, the method further includes: Obtaining the cumulative number of times the network security device is attacked by network attacks corresponding to each network attack type at each historical moment within the preset period; Based on the quadratic function, according to the cumulative number of attacks corresponding to each historical moment, an attack period function of the network attack of each network attack type is fitted; Based on the attack cycle function, determine the maximum cumulative attack number and the minimum cumulative attack number of network attacks of each network attack type within a preset period, and use the maximum cumulative attack number as the maximum attack number of network attacks of the corresponding network attack type; Calculate the attack volume standard value of each network attack type according to the difference between the maximum cumulative attack number and the minimum cumulative attack number; The attack half-life of each network attack type is calculated based on the attack cycle function and the attack amount standard value, and the attack half-life parameter function is constructed based on the attack half-life and the maximum number of attacks.

[0008] According to a flow data processing method based on active noise reduction provided by the present invention, the comprehensive flow data influence of the network security device at the current moment is calculated based on the maximum attack times and the attack half-life of the network attacks of all network attack types, including: Based on the comprehensive flow data impact formula, according to the maximum attack times and the attack half-life of all network attack types, the comprehensive flow data impact of the network security device at the current moment is calculated, and the comprehensive flow data impact formula is:

[0009] in, Indicates the impact of comprehensive traffic data. Indicates the current moment, Indicates i The maximum number of attacks corresponding to the network attack type, Indicates i The attack half-life corresponding to the network attack type of the network attack, C Indicates that the network security device includes C There are three types of network attacks.

[0010] According to a flow data processing method based on active noise reduction provided by the present invention, when it is determined that the impact of the comprehensive flow data is greater than a preset impact threshold, corresponding active noise reduction processing is performed on the flow data of the network security device, including: In the case where it is determined that the comprehensive flow data impact is greater than a preset impact threshold, determining whether the impact difference between the comprehensive flow data impact and the preset impact threshold is less than a preset impact difference, and if the impact difference is greater than or equal to the preset impact difference, performing a flow data prohibition operation on the network security device; If the impact difference is less than the preset impact difference, a traffic data random blocking operation is performed on the traffic data generated by the corresponding attack source in the network security device, wherein the attack source includes the address information of the sender of the network attack; The method further comprises: When it is determined that the comprehensive flow data impact is less than or equal to the preset impact threshold, a flow data passing operation is performed on the network security device.

[0011] According to a flow data processing method based on active noise reduction provided by the present invention, if the influence difference is less than the preset influence difference, a flow data random blocking operation is performed on the flow data generated by the target attack source in the network security device, including: When it is determined that the impact difference is less than the preset impact difference, obtaining the number of historical attacks of the attack source attacking the network security device in the previous preset period; Calculate the attack period parameter of the attack source within the current preset period according to the historical attack number, the attack half-life parameter function and the maximum attack number; When it is determined that the attack cycle parameter is less than or equal to a preset attack cycle threshold, the blocking probability is calculated according to the difference between the attack cycle parameter and the preset attack cycle threshold, wherein the preset attack cycle threshold is half of the duration corresponding to the preset cycle; Based on the traffic data generated by the attack source, a random number corresponding to each traffic data is generated, and a traffic data prohibition operation is performed on the traffic data whose random number is smaller than the blocking probability.

[0012] According to a flow data processing method based on active noise reduction provided by the present invention, the step of obtaining the alarm information data generated by the network security device at the current moment includes: According to the network attacks launched by different attack sources at the current moment on the network security device, network attack type information corresponding to the network attack, attack source address information, attack time point information within the current preset period, and attack number information corresponding to the attack time point information are obtained; The alarm information data is obtained according to the network attack type information, the attack source address information, the attack time point information and the attack number information.

[0013] The present invention also provides a flow data processing system based on active noise reduction, comprising: A converter module, used to obtain the alarm information data generated by the network security device at the current moment; An analyzer module is used to determine, based on the alarm information data, the maximum number of network attacks corresponding to each network attack type suffered by the network security device within a preset period, and the attack half-life of the network attack corresponding to each network attack type, and calculate the comprehensive traffic data impact of the network security device at the current moment based on the maximum number of network attacks and the attack half-life of all network attack types; A noise reducer module, configured to perform corresponding active noise reduction processing on the traffic data of the network security device when determining that the impact of the comprehensive traffic data is greater than a preset impact threshold; Among them, the maximum number of attacks is obtained based on the number of attacks on the network security device at each historical moment within a preset period; the attack half-life represents the length of time required for the number of attacks to reach half of the corresponding maximum number of attacks when the network security device is attacked by a network within a preset period.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a flow data processing method based on active noise reduction as described above is implemented.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the flow data processing method based on active noise reduction as described in any one of the above is implemented.

[0016] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned methods for processing flow data based on active noise reduction.

[0017] The traffic data processing method and system based on active noise reduction provided by the present invention calculate the comprehensive traffic data impact of the network security device at the current moment through the maximum number of network attacks and the attack half-life of all network attack types suffered by the network security device within a preset period, and then, when it is determined that the comprehensive traffic data impact is greater than a preset impact threshold, perform corresponding active noise reduction processing on the traffic data of the network security device, thereby reducing the alarm volume of the network security device and accurately identifying the real threat behavior of the current security device. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A flow chart of a flow data processing method based on active noise reduction provided by the present invention; Figure 2 A curve graph corresponding to the attack half-life parameter function provided by the present invention; Figure 3 A schematic diagram of the structure of a flow data processing system based on active noise reduction provided by the present invention; Figure 4 The overall architecture diagram of the traffic data processing system based on active noise reduction provided by the present invention; Figure 5 This is a schematic structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.

[0021] In the existing optimization technology of network security equipment alarm, the main focus is on optimizing the detection results, but there is a problem. Whether it is a new model, an optimized model, or adding real-time analysis tools, intelligent management tools, etc., there is no significant change in the alarm data of the original equipment. In fact, the addition of new equipment or analysis tools increases the probability of alarm data and repeated alarms, which further increases the number of subsequent analysis tasks. For example, in the prior art, the alarm concentrator is optimized using a machine learning algorithm. Although the alarm is processed using a machine learning method, the alarm volume of the original network security equipment is still not changed. In particular, when a new alarm is added to the alarm concentrator, it is necessary to check not only the alarm in the alarm concentrator, but also the alarm of the original network equipment when performing alarm analysis, resulting in an increase in the workload and false alarm rate.

[0022] At the same time, existing real-time traffic analysis and deep inspection methods may negatively impact network performance and introduce latency, which is a challenge for applications that require high performance and low latency.

[0023] In view of the problems existing in the above-mentioned prior art, the present invention aims at reducing the number of equipment alarms, performs active noise reduction on the original alarm data based on multi-party secure computing, and processes the traffic data without generating new alarm data. The present invention mainly achieves two technical breakthroughs: First, the threat level suffered by the current security device is judged in combination with the attack half-life, and then according to the judgment result, the traffic is blocked through the active noise reduction process, thereby further reducing the number of alarms and achieving the effect of reducing the number of alarms. Secondly, the active noise reduction process only requires simple comparison and judgment, and no training operation is performed, which has no obvious impact on network performance. It should be noted that the traffic data processing method based on active noise reduction provided by the present invention can be applied to a variety of network security scenarios, such as the Internet of Things and cloud security services. The present invention does not specifically limit the application scenarios.

[0024] Figure 1 A flow chart of a method for processing flow data based on active noise reduction provided by the present invention is shown in FIG. Figure 1 As shown, the present invention provides a flow data processing method based on active noise reduction, comprising: Step 101: Acquire the alarm information data generated by the network security device at the current moment.

[0025] In the present invention, each network security device will generate an alarm log, the main contents of which include: alarm name, source address, destination address, start time, end time, alarm number, etc. The present invention collects the required data from the information in these alarm logs, and then processes these data in the corresponding data format to obtain the alarm information data generated by the network security device at the current moment.

[0026] On the basis of the above embodiment, the step of obtaining the alarm information data generated by the network security device at the current moment includes: According to the network attacks launched by different attack sources at the current moment on the network security device, network attack type information corresponding to the network attack, attack source address information, attack time point information within the current preset period, and attack number information corresponding to the attack time point information are obtained; The alarm information data is obtained according to the network attack type information, the attack source address information, the attack time point information and the attack number information.

[0027] Specifically, in the present invention, the alarm information data mainly includes network attack type information, attack source address information, attack time point information and attack number information, among which the network attack type information includes but is not limited to DDoS (distributed denial of service) attacks, SQL injections, cross-site scripting (XSS), malware propagation, port scanning, etc.

[0028] Attack source address information is key information for locating the source of the attack, including IP address, domain name (if available), and possible geographic location information.

[0029] The attack time point information refers to the exact time or time range when the attack occurs. In the present invention, this information mainly represents the time information corresponding to when the network security device is attacked by the network within a preset period. For example, when the preset period is 60 unit times, one of the attack time point information is the 5th unit time of the network attack.

[0030] The attack count information refers to the number of attack attempts of the same type or from the same source address within a preset period.

[0031] Step 102: Determine, based on the alarm information data, the maximum number of network attacks corresponding to each network attack type suffered by the network security device within a preset period, as well as the attack half-life of the network attack corresponding to each network attack type; and calculate the comprehensive traffic data impact of the network security device at the current moment based on the maximum number of network attacks and the attack half-life of all network attack types.

[0032] In the present invention, the network attack behavior cycle is analyzed, and the attack half-life is used as an important parameter to process the traffic data.

[0033] Half-life (symbol t½) is the time required for the amount of (matter) to decrease to half of its initial value. This term is often used in nuclear physics to describe the speed at which unstable atoms undergo radioactive decay or the time that stable atoms survive. In the present invention, the attack half-life parameter function provided is: ; in, y Indicates the current impact level (for example, when the maximum number of attacks within a preset period is 200, it corresponds to the maximum impact level. When the impact level of the attack on the system drops to half of the maximum impact level, the current impact level is 100). Max The maximum number of attacks. T is the attack half-life, t is the current time (e.g., the preset period is 60 units of time, the 4th unit of time). This formula means that the time that has passed in a preset period T When the attack is halved, the impact of a certain type of attack on the system is reduced to half of its original value.

[0034] In the present invention, the attack half-life is related to two characteristics of the attack behavior, one is the attack intensity, and the other is the duration. When calculating the attack half-life, historical related data is used to analyze the number of attacks per unit time corresponding to the attack source IP, so as to obtain the maximum number of attacks for each attack type, and then the corresponding attack half-life is calculated by the above attack half-life parameter function.

[0035] Step 103, when it is determined that the impact of the comprehensive traffic data is greater than a preset impact threshold, performing corresponding active noise reduction processing on the traffic data of the network security device; Among them, the maximum number of attacks is obtained based on the number of attacks on the network security device at each historical moment within a preset period; the attack half-life represents the length of time required for the number of attacks to reach half of the corresponding maximum number of attacks when the network security device is attacked by a network within a preset period.

[0036] In the present invention, when it is determined that the impact of traffic data is greater than a preset threshold, active noise reduction processing measures will be taken to intervene or process the traffic data that may have a negative impact, so as to reduce its potential risk or impact. In the present invention, the specific method of active noise reduction processing is mainly achieved through blocking. For example, when the threat level is low, all traffic is released; when the threat level is medium, traffic is released randomly; when the threat level is high, traffic is not released.

[0037] The traffic data processing method based on active noise reduction provided by the present invention calculates the comprehensive traffic data influence of the network security device at the current moment through the maximum number of network attacks and the attack half-life of all network attack types suffered by the network security device within a preset period, and then, when it is determined that the comprehensive traffic data influence is greater than a preset influence threshold, performs corresponding active noise reduction processing on the traffic data of the network security device, thereby reducing the alarm volume of the network security device and accurately identifying the real threat behavior of the current security device.

[0038] Based on the above embodiment, the method further includes: Obtaining the cumulative number of times the network security device is attacked by network attacks corresponding to each network attack type at each historical moment within the preset period; Based on the quadratic function, according to the cumulative number of attacks corresponding to each historical moment, an attack period function of the network attack of each network attack type is fitted; Based on the attack cycle function, determine the maximum cumulative attack number and the minimum cumulative attack number of network attacks of each network attack type within a preset period, and use the maximum cumulative attack number as the maximum attack number of network attacks of the corresponding network attack type; Calculate the attack volume standard value of each network attack type according to the difference between the maximum cumulative attack number and the minimum cumulative attack number; The attack half-life of each network attack type is calculated based on the attack cycle function and the attack amount standard value, and the attack half-life parameter function is constructed based on the attack half-life and the maximum number of attacks.

[0039] Each network security device will generate an alarm log, the main contents of which include the alarm name, source address, destination address, start time, end time and alarm number. In the present invention, based on the data required for calculating the attack half-life, these alarm logs are first converted and processed to obtain the data format shown in Table 1: Table 1 Alarm data conversion results

[0040] In the present invention, timing is performed by broadcasting a clock. For example, the frequency is set to one minute by default, that is, a broadcast is performed once every minute to synchronously collect the half-life calculation data within each preset period. The specific steps are: In the initialization phase, the initial time is set to t0. In the present invention, the broadcast clock parameter t is set as the unit time, which is 1 minute by default. t0 reaches the time t1 after time t. 60 unit times are set as one attack cycle, that is, one attack cycle is 1 hour, so the attack half-life is calculated within one attack cycle.

[0041] Furthermore, every time a unit of time passes, a signal sent by the broadcast clock is obtained, and based on each unit of time, the attack information corresponding to the network security device (such as the data structure shown in Table 1) is summarized.

[0042] Furthermore, after an attack cycle, such as 60 minutes, the existing attack information is processed and analyzed to calculate the number of attacks for each attack type. For example, after summarizing, the attack types of nmap scanning can be obtained as shown in Table 2: Table 2 Attack data summary

[0043] In Table 2, each piece of data represents a data format that needs to be output. For example, Converter 01 may identify multiple attacks, including nmap scan attacks against 192.168.0.22 and 192.168.0.23, with the number of attacks being 86 and 56.

[0044] Next, the attack cycle is calculated based on the attack type, attack source IP, time point, and attack times. For example, the attack source IP is 192.168.0.22, and the attack behavior on Converter 01 is 86 times at time point t1; the attack behavior on Converter 02 is 77 times at time point t5; 10 At that moment, the number of attacks added up to 60. Similarly, the number of attacks by the attack source IP at all time points within a preset period can be obtained, and the summary results are shown in Table 3: Table 3 Summary of attack times of a certain attack source IP

[0045] In the present invention, the time point is regarded as the x-coordinate, the accumulated value of the number of attacks is regarded as the y-coordinate, and a quadratic function is used for fitting. According to the above Table 3, the relevant coordinates can be obtained: The cumulative value of the number of attacks corresponding to time point t1 is 86, and the coordinates are (1, 86); The cumulative number of attacks corresponding to time point t5 is 86+77=163, and the coordinates are (5, 163); Time point t 10 The corresponding cumulative number of attacks is 163+60=223, and the coordinates are (10, 223); Time point t 18 The corresponding cumulative number of attacks is 223+93=316, and the coordinates are (18, 316).

[0046] Using the quadratic function y=ax 2 +bx+c fits the above coordinates and obtains: a=-0.2690456135602434; b = 18.409440420189956; c = 70.78738795391274; Finally, the attack cycle function of this network attack type is obtained as follows: y=﹣0.27x 2 +18.41x+70.79.

[0047] Correspondingly, after the above calculation process, we can get the attack cycle function of various types of attacks on IP 192.168.0.22. For example, nmap scan: y=a1x 2 +b1x+c1; brute force cracking: y=a2x 2 +b2x+c2; SQL injection: y=a3x 2 +b3x+c3 and other attack types. It should be noted that for any attack source IP, the IP does not necessarily involve all types of attacks, and only the attack types included in the IP are counted.

[0048] Finally, calculate the time corresponding to the standard value of the attack volume in each attack cycle function. For example, for the NAMP scan, its attack cycle function is y = -0.27x 2 +18.41x+70.79. Since the attack times are fitted in a cumulative manner, the function is an increasing function. Therefore, during the attack period, y=-0.27x 2 The minimum value of +18.41x+70.79 appears at x=0, and the maximum value appears at x=60. Then, we find the values ​​corresponding to these two points, and subtract the minimum value from the maximum value to get the difference in the number of attacks. That is, when x=0, the minimum value of y is 70.79; when x=60, the maximum value of y is 203.39, so the standard value of the attack volume is (203.39-70.79) / 2+70.79=137.09.

[0049] Then, based on the attack period function, and considering that the value of x should be within the domain, when y=137.09, the corresponding x value is 3.81. Therefore, when x=3.81, the standard value of the attack amount is obtained, and the value of the attack half-life T is 3.81≈4.

[0050] In one embodiment, according to the maximum number of attacks of 203.39≈203, the influence of the attack on the system should be calculated according to the following steps. In the current attack cycle, the attack behavior of a certain IP is modeled and analyzed, and the maximum impact is combined with the attack half-life to determine the impact of the attack on the system. If the attack half-life is 4, then at time 4, the number of attacks should be reduced to half of the maximum number of attacks in 60 cycles, 203, and the impact of the network attack on the system will also be reduced to half of the maximum impact. For example, for the above-mentioned attack source IP of 192.168.0.22, the attack half-life parameter function corresponding to the influence of the nmap attack type is: .

[0051] As time goes by, the impact of nmap attacks on the system becomes less and less. Through the above attack half-life calculation process, the attack half-life corresponding to each type of network attack is obtained, as shown in Table 4: Table 4 Attack half-life

[0052] On the basis of the above embodiment, the comprehensive flow data influence of the network security device at the current moment is calculated based on the maximum attack times and the attack half-life of the network attacks of all network attack types, including: Based on the comprehensive flow data impact formula, according to the maximum attack times and the attack half-life of all network attack types, the comprehensive flow data impact of the network security device at the current moment is calculated, and the comprehensive flow data impact formula is:

[0053] in, Indicates the impact of comprehensive traffic data. Indicates the current moment, Indicates i The maximum number of attacks corresponding to the network attack type, Indicates i The attack half-life corresponding to the network attack type of the network attack, C Indicates that the network security device includes C There are three types of network attacks.

[0054] In the present invention, the alarm log is generated according to the network security device, and the specific data format constructed is shown in Table 5: Table 5 Data transmission of network security devices

[0055] Through the data in Table 5 above, the maximum number of attacks on a certain network security device within a preset period can be obtained. , then the comprehensive traffic data impact at the current time Influence now (Abbreviation IF now ) is calculated as follows:

[0056] In the present invention, Represents the half-life of a certain type of network attack. Assuming that a total of C This attack, by all C The effects of these attacks are superimposed (that is, from i =1 to C ). Indicates the current network security device i The maximum number of attacks of the same type of network attack in a cycle (the default is 60 units of time, that is, 1 hour). Then, based on the calculated comprehensive traffic data impact , determine the current network security situation of the network security device, and handle the traffic processed by the security device according to the network security situation.

[0057] The present invention comprehensively determines an impact result based on all network attack types, and then performs unified operations on all network attack types based on the impact result. The specific process is to calculate the impact of all network attack types and superimpose the impact results of all network attack types to obtain the current comprehensive traffic data impact degree. , judge the threat status of the current network security equipment according to the impact value, and judge the impact of comprehensive traffic data by setting thresholds High school and low levels.

[0058] On the basis of the above embodiment, when it is determined that the impact of the comprehensive traffic data is greater than a preset impact threshold, the corresponding active noise reduction processing is performed on the traffic data of the network security device, including: In the case where it is determined that the comprehensive flow data impact is greater than a preset impact threshold, determining whether the impact difference between the comprehensive flow data impact and the preset impact threshold is less than a preset impact difference, and if the impact difference is greater than or equal to the preset impact difference, performing a flow data prohibition operation on the network security device; If the impact difference is less than the preset impact difference, a traffic data random blocking operation is performed on the traffic data generated by the corresponding attack source in the network security device, wherein the attack source includes the address information of the sender of the network attack; The method further comprises: When it is determined that the comprehensive flow data impact is less than or equal to the preset impact threshold, a flow data passing operation is performed on the network security device.

[0059] In the present invention, the preset impact threshold can be set to 50% of the historical maximum attack times by default. If the threshold is exceeded, a noise reduction operation is performed. Preferably, in one embodiment, after running for a period of time, the threshold can be updated according to the impact of each comprehensive flow data (such as adjusting according to the impact ratio of the comprehensive flow data between two preset periods).

[0060] In the present invention, the active noise reduction process is to filter the network device traffic, and the active noise reduction strategy can be: when the threat level is low, such as when the comprehensive traffic data impact is less than or equal to the preset impact threshold, all traffic is released; when the threat level is medium, such as when the comprehensive traffic data impact is greater than the preset impact threshold, and the difference between the two is less than the preset impact difference, the traffic is randomly released; when the threat level is high, such as when the comprehensive traffic data impact is greater than the preset impact threshold, and the difference between the two is greater than the preset impact difference, the traffic is not released. In the present invention, the comprehensive traffic data impact obtained according to the attack half-life is an ideal, general state of threat level change result, which can be used as a benchmark to judge the attack behavior suffered by the current security device A, thereby randomly blocking the traffic of a certain attack source IP.

[0061] On the basis of the above embodiment, if the impact difference is less than the preset impact difference, performing a traffic data random blocking operation on the traffic data generated by the target attack source in the network security device includes: When it is determined that the impact difference is less than the preset impact difference, obtaining the number of historical attacks of the attack source attacking the network security device in the previous preset period; Calculate the attack period parameter of the attack source within the current preset period according to the historical attack number, the attack half-life parameter function and the maximum attack number; When it is determined that the attack cycle parameter is less than or equal to a preset attack cycle threshold, the blocking probability is calculated according to the difference between the attack cycle parameter and the preset attack cycle threshold, wherein the preset attack cycle threshold is half of the duration corresponding to the preset cycle; Based on the traffic data generated by the attack source, a random number corresponding to each traffic data is generated, and a traffic data prohibition operation is performed on the traffic data whose random number is smaller than the blocking probability.

[0062] The existing processing methods for attack source IP only have two states: "ban" and "pass". The present invention proposes a random filtering processing method, which is an optimization of the two existing processing methods, so that the request flow of a certain attack source IP is dynamically adjusted according to its threat level.

[0063] Specifically, in one embodiment, based on the current time, the number of nmap scanning attacks of IP: 192.168.0.22 in the previous attack cycle (60 units of time by default) is first calculated, so as to determine the attack status of the nmap scanning process of this IP. Assuming that the number of historical attacks in the previous attack cycle is 500, and the number of attacks in the current attack cycle is 203, substitute them into the corresponding attack half-life parameter function: ; Get the value x1 of the corresponding attack cycle parameter, Figure 2 The curve diagram corresponding to the attack half-life parameter function provided by the present invention can be specifically referred to Figure 2 shown.

[0064] Further, in the present invention, according to the obtained attack cycle parameters, there are two situations: the first situation is that the attack cycle parameter x1 is Figure 2 The left side of x=30 (based on half a cycle, the current attack cycle is 60, so half is 30) indicates that there is a high threat risk and needs to be blocked.

[0065] Another case is that the attack period parameter x1 is Figure 2 The right side of x=30 in ( Figure 2 ), indicating that the attack is in its final stage, or has not yet posed a substantial threat, and thus may not be blocked at this time.

[0066] Furthermore, in the present invention, when the time point corresponding to the attack cycle parameter x1 has not reached half of the current cycle, the specific blocking process is related to the distance from half of the attack cycle. The longer the distance, the higher the blocking rate. Taking the preset cycle of 60 as an example, the calculation method of the blocking probability P is: ; If x1=8, then P=0.22=22%.

[0067] Furthermore, the blocking process in the present invention discards each flow based on the generated random number, generates a random number of random (0, 1) for each request flow, and determines the size of the random number corresponding to each flow and the blocking probability P. If it is less than P, the flow is blocked and needs to be blocked; if it is greater than P, the flow passes normally. Through the above blocking process, if the cumulative number of attacks on 192.168.0.22 per unit time is more, the corresponding x value is smaller, and the farther it is from half of the preset period, the higher the blocking probability is, such as the maximum blocking rate of 30% in the above embodiment.

[0068] The present invention can protect network security devices by blocking the traffic of abnormal attack source IPs, fundamentally reducing the number of alarms of a certain network security device and achieving the effect of active noise reduction.

[0069] The flow data processing system based on active noise reduction provided by the present invention is described below. The flow data processing system based on active noise reduction described below and the flow data processing method based on active noise reduction described above can correspond to each other.

[0070] Figure 3 The structure diagram of the flow data processing system based on active noise reduction provided by the present invention is as follows: Figure 3 As shown, the present invention provides a flow data processing system based on active noise reduction, including a converter module 301, an analyzer module 302 and a noise reducer module 303, wherein the converter module 301 is used to obtain the alarm information data generated by the network security device at the current moment; the analyzer module 302 is used to determine the maximum number of attacks corresponding to each network attack type suffered by the network security device within a preset period, and the attack half-life of each network attack type according to the alarm information data, and calculate the comprehensive flow data influence of the network security device at the current moment according to the maximum number of attacks and the attack half-life of the network attacks of all network attack types; the noise reducer module 303 is used to perform corresponding active noise reduction processing on the flow data of the network security device when it is determined that the comprehensive flow data influence is greater than a preset influence threshold; wherein the maximum number of attacks is obtained according to the number of attacks suffered by the network security device at each historical moment within a preset period; the attack half-life represents the length of time required for the number of attacks to reach half of the corresponding maximum number of attacks when the network security device is attacked within a preset period.

[0071] In the present invention, the input end of the noise reducer module is connected to the output end of the router switch and other devices, and the output end of the noise reducer module is connected to the input end of the network security device (such as IDS, WAF and IPS, etc.). The main function of the noise reducer module is to undertake the task of filtering network traffic, generate corresponding dynamic condition parameters based on the comprehensive traffic data influence obtained by the analyzer module, establish different filtering models, and realize active noise reduction of attack traffic.

[0072] The input end of the converter module is connected to the output end of the network security device, and is used to collect and process the data (alarm log) intercepted by the network security device. These data record information such as the attack source IP, destination IP, suspected attack type, and attack time. The main function of the converter module is to convert the format of these data and then input them into the analyzer module for analysis and processing.

[0073] The input end of the analyzer module is connected to the output end of the converter module and is connected to the denoiser module. The analyzer module passes the calculated half-life parameters of all attack types to the denoiser module for modeling analysis, and the network security device passes the traffic and label information to the denoiser module. The analyzer module analyzes the alarm information transmitted by the converter module, obtains the relevant denoising parameters, and sends the denoising parameters to the denoiser module for deployment. The denoiser module makes a security status judgment based on the half-life parameters sent by the analyzer module, the traffic, label information and maximum attack value sent by the network security device, and actively intercepts abnormal traffic to achieve the effect of denoising. Figure 4 This is the overall architecture diagram of the traffic data processing system based on active noise reduction provided by the present invention. The deployment methods of the noise reducer, converter, analyzer and other modules can be referred to Figure 4 shown.

[0074] In the present invention, the concept of attack half-life is proposed, and the attack half-life is used as an important parameter to process the traffic data. According to the attack half-life and analysis, the maximum number of attacks of each network attack type is obtained, and then the half-life parameters of all network attack types are obtained, so as to analyze the attack behavior cycle, thereby judging the security status of the current attack behavior according to the half-life parameters and realizing active noise reduction.

[0075] In the process of active noise reduction, the threat to the network security device is mainly judged based on the half-life function, the attack distribution state, and the attack duration, and the network traffic is blocked based on whether it exceeds the threat threshold. The threat judgment is mainly related to the attack intensity. The higher the deviation between the threatening IP traffic and the normal traffic, the stronger the blocking force. The present invention realizes active noise reduction of traffic data by actively processing the network traffic, which can reduce the number of equipment alarms, and accurately identify the real threat behavior of the current security device in combination with the attack half-life.

[0076] The traffic data processing system based on active noise reduction provided by the present invention calculates the comprehensive traffic data impact of the network security device at the current moment through the maximum number of attacks and the attack half-life of all network attack types suffered by the network security device within a preset period, and then, when it is determined that the comprehensive traffic data impact is greater than a preset impact threshold, performs corresponding active noise reduction processing on the traffic data of the network security device, thereby reducing the alarm volume of the network security device and accurately identifying the real threat behavior of the current security device.

[0077] The system provided by the present invention is used to execute the above-mentioned method embodiments. Please refer to the above-mentioned embodiments for the specific processes and detailed contents, which will not be repeated here.

[0078] Figure 5 A schematic diagram of the structure of an electronic device provided by the present invention, such as Figure 5 As shown, the electronic device may include: a processor (Processor) 501, a communication interface (Communications Interface) 502, a memory (Memory) 503 and a communication bus 504, wherein the processor 501, the communication interface 502, and the memory 503 communicate with each other through the communication bus 504. The processor 501 can call the logic instructions in the memory 503 to execute the traffic data processing method based on active noise reduction, which includes: obtaining the alarm information data generated by the network security device at the current moment; determining the maximum number of network attacks corresponding to each network attack type suffered by the network security device within a preset period according to the alarm information data, and the attack half-life of the network attack corresponding to each network attack type, and calculating the comprehensive traffic data influence of the network security device at the current moment according to the maximum number of attacks and the attack half-life of the network attacks of all network attack types; when it is determined that the comprehensive traffic data influence is greater than the preset influence threshold, performing corresponding active noise reduction processing on the traffic data of the network security device; wherein the maximum number of attacks is obtained according to the number of attacks suffered by the network security device at each historical moment within a preset period; the attack half-life represents the length of time required for the number of attacks to reach half of the corresponding maximum number of attacks when the network security device is attacked within a preset period.

[0079] In addition, the logic instructions in the above-mentioned memory 503 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0080] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the traffic data processing method based on active noise reduction provided by the above methods, the method comprising: obtaining alarm information data generated by a network security device at the current moment; determining, according to the alarm information data, the maximum number of attacks corresponding to each network attack type suffered by the network security device within a preset period, and the attack half-life of each network attack type corresponding to the network attack, and calculating the comprehensive traffic data influence of the network security device at the current moment according to the maximum number of attacks and the attack half-life of the network attacks of all network attack types; when it is determined that the comprehensive traffic data influence is greater than a preset influence threshold, performing corresponding active noise reduction processing on the traffic data of the network security device; wherein the maximum number of attacks is obtained according to the number of attacks suffered by the network security device at each historical moment within a preset period; the attack half-life represents the length of time required for the number of attacks to reach half of the corresponding maximum number of attacks when the network security device is attacked within a preset period.

[0081] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the processor executes the traffic data processing method based on active noise reduction provided in the above-mentioned embodiments, the method comprising: obtaining alarm information data generated by the network security device at the current moment; determining, according to the alarm information data, the maximum number of network attacks corresponding to each network attack type suffered by the network security device within a preset period, and the attack half-life of the network attack corresponding to each network attack type, and calculating the comprehensive traffic data influence of the network security device at the current moment according to the maximum number of attacks and the attack half-life of the network attacks of all network attack types; when it is determined that the comprehensive traffic data influence is greater than a preset influence threshold, performing corresponding active noise reduction processing on the traffic data of the network security device; wherein the maximum number of attacks is obtained according to the number of attacks suffered by the network security device at each historical moment within a preset period; the attack half-life represents the length of time required for the number of attacks to reach half of the corresponding maximum number of attacks when the network security device is attacked within a preset period.

[0082] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0083] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A flow data processing method based on active noise reduction, characterized in that: include: Obtain the alarm information data generated by the network security device at the current moment; Determine, based on the alarm information data, the maximum number of network attacks corresponding to each network attack type suffered by the network security device within a preset period, and the attack half-life of each network attack type corresponding to the network attack, and calculate the comprehensive traffic data impact of the network security device at the current moment based on the maximum number of attacks and the attack half-life of network attacks of all network attack types; When it is determined that the impact of the comprehensive traffic data is greater than a preset impact threshold, performing corresponding active noise reduction processing on the traffic data of the network security device; Among them, the maximum number of attacks is obtained based on the number of attacks on the network security device at each historical moment within a preset period; the attack half-life represents the length of time required for the number of attacks to reach half of the corresponding maximum number of attacks when the network security device is attacked by a network within a preset period.

2. The method for processing flow data based on active noise reduction according to claim 1, characterized in that: The method further comprises: Obtaining the cumulative number of times the network security device is attacked by network attacks corresponding to each network attack type at each historical moment within the preset period; Based on the quadratic function, according to the cumulative number of attacks corresponding to each historical moment, an attack period function of the network attack of each network attack type is fitted; Based on the attack cycle function, determine the maximum cumulative attack number and the minimum cumulative attack number of network attacks of each network attack type within a preset period, and use the maximum cumulative attack number as the maximum attack number of network attacks of the corresponding network attack type; Calculate the attack volume standard value of each network attack type according to the difference between the maximum cumulative attack number and the minimum cumulative attack number; The attack half-life of each network attack type is calculated based on the attack cycle function and the attack amount standard value, and the attack half-life parameter function is constructed based on the attack half-life and the maximum number of attacks.

3. The method for processing flow data based on active noise reduction according to claim 2, characterized in that: The calculating, based on the maximum attack times and the attack half-life of the network attacks of all network attack types, the comprehensive flow data influence of the network security device at the current moment comprises: Based on the comprehensive flow data impact formula, according to the maximum attack times and the attack half-life of all network attack types, the comprehensive flow data impact of the network security device at the current moment is calculated, and the comprehensive flow data impact formula is: in, Indicates the impact of comprehensive traffic data. Indicates the current moment, Indicates i The maximum number of attacks corresponding to the network attack type, Indicates i The attack half-life corresponding to the network attack type of the network attack, C Indicates that the network security device includes C There are three types of network attacks.

4. The method for processing flow data based on active noise reduction according to claim 2, characterized in that: When it is determined that the impact of the comprehensive traffic data is greater than a preset impact threshold, performing corresponding active noise reduction processing on the traffic data of the network security device includes: In the case where it is determined that the comprehensive flow data impact is greater than a preset impact threshold, determining whether the impact difference between the comprehensive flow data impact and the preset impact threshold is less than a preset impact difference, and if the impact difference is greater than or equal to the preset impact difference, performing a flow data prohibition operation on the network security device; If the impact difference is less than the preset impact difference, a traffic data random blocking operation is performed on the traffic data generated by the corresponding attack source in the network security device, wherein the attack source includes the address information of the sender of the network attack; The method further comprises: When it is determined that the comprehensive flow data impact is less than or equal to the preset impact threshold, a flow data passing operation is performed on the network security device.

5. The method for processing flow data based on active noise reduction according to claim 4, characterized in that: If the impact difference is less than the preset impact difference, performing a traffic data random blocking operation on the traffic data generated by the target attack source in the network security device includes: When it is determined that the impact difference is less than the preset impact difference, obtaining the number of historical attacks of the attack source attacking the network security device in the previous preset period; Calculate the attack period parameter of the attack source within the current preset period according to the historical attack number, the attack half-life parameter function and the maximum attack number; When it is determined that the attack cycle parameter is less than or equal to a preset attack cycle threshold, the blocking probability is calculated according to the difference between the attack cycle parameter and the preset attack cycle threshold, wherein the preset attack cycle threshold is half of the duration corresponding to the preset cycle; Based on the traffic data generated by the attack source, a random number corresponding to each traffic data is generated, and a traffic data prohibition operation is performed on the traffic data whose random number is smaller than the blocking probability.

6. The method for processing flow data based on active noise reduction according to claim 1, characterized in that: The obtaining of the alarm information data generated by the network security device at the current moment includes: According to the network attacks launched by different attack sources at the current moment on the network security device, network attack type information corresponding to the network attack, attack source address information, attack time point information within the current preset period, and attack number information corresponding to the attack time point information are obtained; The alarm information data is obtained according to the network attack type information, the attack source address information, the attack time point information and the attack number information.

7. A flow data processing system based on active noise reduction, characterized in that: include: A converter module, used to obtain the alarm information data generated by the network security device at the current moment; An analyzer module is used to determine, based on the alarm information data, the maximum number of network attacks corresponding to each network attack type suffered by the network security device within a preset period, and the attack half-life of the network attack corresponding to each network attack type, and calculate the comprehensive traffic data impact of the network security device at the current moment based on the maximum number of network attacks and the attack half-life of all network attack types; A noise reducer module, configured to perform corresponding active noise reduction processing on the traffic data of the network security device when determining that the impact of the comprehensive traffic data is greater than a preset impact threshold; Among them, the maximum number of attacks is obtained based on the number of attacks on the network security device at each historical moment within a preset period; the attack half-life represents the length of time required for the number of attacks to reach half of the corresponding maximum number of attacks when the network security device is attacked by a network within a preset period.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the flow data processing method based on active noise reduction as claimed in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for processing traffic data based on active noise reduction according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for processing traffic data based on active noise reduction according to any one of claims 1 to 6 is implemented.

Citation Information

Patent Citations

  • Alarm reduction method and device, equipment and computer readable storage medium

    CN113315785A

  • Network alarm information clustering method based on attribute correlation

    CN113569116A

  • Noise reduction method and device for Web attack

    CN117135026A

  • Alarm noise reduction method based on time domain and graph domain characteristics in alarm storm period

    CN117395127A

  • Whole-flow processing system and method based on network security alarm event

    CN117614717A