Network behavior monitoring and positioning method and system
Through AI analysis and edge computing, combined with machine learning and deception technology, the abnormal behavior of networks is identified and handled, and the problems of insufficient detection accuracy and slow response speed in the existing technology are solved, and more efficient network security monitoring and response are achieved.
Patent Information
- Application Number
- CN202411319873.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-20
- Publication Date
- 2024-12-03
AI Technical Summary
The existing network security monitoring technology is insufficient in detecting complex network behavior patterns, poor adaptability to new attack methods, difficult to achieve real-time monitoring and rapid response, and lacks refined and personalized abnormal behavior handling strategies.
AI analysis and judgment technology is used to analyze network traffic data, combine machine learning algorithms to establish a behavior pattern library, conduct real-time analysis through edge computing nodes, use traffic feature quantity and deception technology to detect and identify abnormal behaviors, and mark and manage them through multi-dimensional analysis.
It significantly improves the recognition accuracy and response speed of network abnormal behavior, can quickly discover and deal with potential security threats, and provides more refined and intelligent network security protection.
Smart Images

Figure CN119071079A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of network security monitoring, and more specifically, to a network behavior monitoring and positioning method and system. Background Art
[0002] Existing network security monitoring technologies mainly rely on traditional traffic analysis and intrusion detection systems, which usually identify abnormal behaviors based on rules and signatures. However, with the continuous evolution and complexity of network attack methods, traditional methods have shown obvious limitations in detecting unknown attacks and advanced persistent threats (APTs). In addition, existing technologies often have performance bottlenecks when processing large-scale data, making it difficult to achieve real-time monitoring and rapid response. In terms of network behavior analysis, there is also a lack of sufficient intelligence and automation, making it difficult to accurately identify and locate abnormal behaviors from massive data.
[0003] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the prior art: the existing monitoring system lacks accuracy in identifying complex network behavior patterns and has poor adaptability to new attack methods; in large-scale network environments, the ability of real-time analysis and response is limited, and it is difficult to meet the needs of modern network environments for rapid detection and disposal; at the same time, the prior art lacks refined and personalized processing strategies when marking and managing abnormal behaviors, and is unable to effectively distinguish different types of abnormal behaviors and take corresponding security measures. Summary of the invention
[0004] The invention provides a network behavior monitoring and positioning method and system.
[0005] In a first aspect of the present invention, a network behavior monitoring and positioning method is provided, comprising:
[0006] Obtain traffic data of each node in the network;
[0007] Use AI analysis technology to analyze traffic data and determine whether there is abnormal behavior;
[0008] If it does not exist, the network behavior is normal. If it does exist, the abnormal behavior characteristics and their types are recorded. In the abnormal behavior analyzed, if the traffic characteristic quantity is higher than the set threshold, it is an abnormal signal and is marked as a blacklist. If the traffic characteristic quantity is equal to the set threshold, it is a critical state signal and is marked as a whitelist. If the traffic characteristic quantity is less than the set threshold, it is a normal state signal and is not marked.
[0009] If AI identifies a malicious behavior pattern, which is an abnormal state signal, it will be marked as blacklisted; if AI does not identify a malicious behavior pattern, which is a normal state signal, it will not be marked;
[0010] If an abnormal message is found through packet capture analysis of the dedicated network card of the edge computing node, it is an abnormal state signal, and it is marked as a blacklist; if no abnormal message is found, it is a normal state signal, and it is not marked;
[0011] If the deception technology detects malicious detection behavior, which is an abnormal state signal, it will be marked as blacklisted; if no malicious detection behavior is detected, which is a normal state signal, it will not be marked;
[0012] Based on the comparison results of the traffic feature F and the set threshold, as well as the results of AI analysis, packet capture analysis, and deception technology detection, the abnormal indicators a, b, c, and d of the network behavior are determined, where:
[0013] If F is higher than the set threshold, a=1; if it is equal to, a=0; if it is lower, a=-1;
[0014] If AI identifies a malicious behavior pattern with a matching degree exceeding the preset threshold, b = 1; if not, b = 0;
[0015] If an abnormal message is found through packet capture analysis by the dedicated network card, c = 1; if no abnormal message is found, c = 0;
[0016] If the deception technology detects malicious probing behavior, then d = 1; if not, then d = 0;
[0017] If a, b, c, and d are all less than or equal to 1, the network behavior is normal; if one or two of a, b, c, and d are 1 and the remaining values are -1, it is a suspected abnormal state and the next set of data is needed to determine whether abnormal characteristics appear;
[0018] If one or two of a, b, c, and d are 1 and the remaining value is 0, the network behavior is in a critical abnormal state, and a warning signal is output and the network behavior is continuously monitored; if three or more of a, b, c, and d are 1, the network behavior is in an abnormal state, and a warning signal is output and an alarm is promptly sent to the security management center. If a, b, c, and d are all less than or equal to 1, the network behavior is normal; if one or two of a, b, c, and d are 1 and the remaining value is -1, it is a suspected abnormal state, and the next set of data is waited for to determine whether abnormal characteristics appear;
[0019] If one or two of a, b, c, and d are 1 and the remaining values are 0, the network behavior is in a critical abnormal state, and a warning signal is output and the network behavior is continuously monitored; if three or more of a, b, c, and d are 1, the network behavior is in an abnormal state, and a warning signal is output and an alarm is promptly sent to the security management center.
[0020] Furthermore, when monitoring network traffic, edge computing nodes are used for real-time analysis.
[0021] Furthermore, the method of AI judging and identifying malicious behavior patterns is to train historical traffic data through machine learning algorithms and establish a behavior pattern library. When the degree of match between real-time traffic data and malicious behavior patterns in the library exceeds a preset threshold, it is identified as an abnormal state.
[0022] Furthermore, the preset threshold is a matching degree between 0.85 and 1.
[0023] Furthermore, the WiFi radio messages captured by the dedicated network card are positively correlated with the abnormality of network behavior.
[0024] Furthermore, the malicious probing behaviors detected by the deception technology are positively correlated with the abnormality of network behaviors.
[0025] In a second aspect of the present invention, a network behavior monitoring and positioning system is provided, comprising:
[0026] A data acquisition module is used to obtain the flow data of each node in the network;
[0027] The analysis module is used to analyze traffic data using AI analysis technology to determine whether there is abnormal behavior;
[0028] The marking module is used to mark and manage abnormal behaviors;
[0029] A judgment module is used to judge abnormal characteristics of network behavior;
[0030] A warning module is used to output a warning signal and perform corresponding processing when an abnormal state is detected;
[0031] The security management center is used to receive alarm signals and make further security responses.
[0032] According to the above-mentioned embodiments of the present invention, at least the following beneficial effects are achieved: the network behavior monitoring and positioning method of the present invention can significantly improve the recognition accuracy and response speed of abnormal network behavior. By using AI judgment technology to conduct in-depth analysis of traffic data and combining machine learning algorithms to establish a behavior pattern library, this method can effectively identify real-time traffic data that matches malicious behavior patterns in the library to a degree exceeding a preset threshold, thereby quickly discovering and responding to potential security threats. In addition, this method can reduce data transmission delays and improve monitoring efficiency by performing real-time analysis through edge computing nodes.
[0033] At the same time, this method constructs a comprehensive traffic feature calculation model by comprehensively considering traffic feature quantities in multiple dimensions, such as packet size, transmission rate, network layer height, transmission distance, and average transmission efficiency. This can not only more accurately evaluate the abnormality of network behavior, but also flexibly classify and mark abnormal behaviors by setting different thresholds, including the division of blacklists and whitelists. This multi-dimensional analysis method can provide richer and more detailed decision support for network security management, and help build a more robust and intelligent network security protection system. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, in which:
[0035] Figure 1 A flowchart of a network behavior monitoring and positioning method provided by an embodiment of the present invention;
[0036] Figure 2 A schematic diagram of the structure of a network behavior monitoring and positioning system provided by an embodiment of the present invention;
[0037] Figure 3 The schematic diagram schematically shows the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION
[0038] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.
[0039] Those skilled in the art know that the embodiments of the present invention can be implemented as a system, device, apparatus, method or computer program product. Therefore, the present invention can be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.
[0040] It should be noted that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction and does not have any limiting meaning.
[0041] Reference below Figure 1 , Figure 1 The flowchart of the network behavior monitoring and positioning method provided by one embodiment of the present invention is as follows. Figure 1 As shown, a network behavior monitoring and positioning method 100 includes:
[0042] Step 101, obtaining traffic data of each node in the network;
[0043] Step 102, using AI analysis technology to analyze the traffic data to determine whether there is abnormal behavior;
[0044] Step 103, if it does not exist, the network behavior is normal, if it does exist, the abnormal behavior characteristics and their types are recorded; in the abnormal behavior analyzed, if the flow characteristic quantity is higher than the set threshold, it is an abnormal signal and is marked as a blacklist; if the flow characteristic quantity is equal to the set threshold, it is a critical state signal and is marked as a whitelist; if the flow characteristic quantity is less than the set threshold, it is a normal state signal and is not marked;
[0045] Step 104: If the AI identifies a malicious behavior pattern and it is an abnormal state signal, it is marked as a blacklist; if the AI does not identify a malicious behavior pattern and it is a normal state signal, it is not marked;
[0046] Step 105: If an abnormal message is found through packet capture analysis of the dedicated network card of the edge computing node, which is an abnormal state signal, it is marked as a blacklist; if no abnormal message is found, it is a normal state signal, and it is not marked;
[0047] Step 106: If the deception technology detects malicious detection behavior, which is an abnormal state signal, it is marked as a blacklist; if no malicious detection behavior is detected, which is a normal state signal, it is not marked;
[0048] The flow characteristic calculation formula is:
[0049]
[0050] Among them, F is the flow characteristic quantity;
[0051] n is the total number of packets in the observation period;
[0052] I i is the traffic intensity of the ith data packet, including the size or transmission rate of the data packet;
[0053] h i is the network level height of the ith data packet, including the number of network layers or routing hops that the data packet passes through;
[0054] z i is the transmission distance of the ith data packet, including the network distance from the source to the destination of the data packet;
[0055] σ is the average transmission efficiency of the network, including the inverse of the average packet loss rate of the network;
[0056] ω is the signal frequency of the network, including the average frequency of data packet transmission in the network.
[0057] Based on the comparison results of the traffic feature F and the set threshold, as well as the results of AI analysis, packet capture analysis, and deception technology detection, the abnormal indicators a, b, c, and d of the network behavior are determined, where:
[0058] If F is higher than the set threshold, a=1; if it is equal to, a=0; if it is lower, a=-1;
[0059] If AI identifies a malicious behavior pattern with a matching degree exceeding the preset threshold, b = 1; if not, b = 0;
[0060] If an abnormal message is found through packet capture analysis by the dedicated network card, c = 1; if no abnormal message is found, c = 0;
[0061] If the deception technology detects malicious probing behavior, then d = 1; if not, then d = 0;
[0062] If a, b, c, and d are all less than or equal to 1, the network behavior is normal; if one or two of a, b, c, and d are 1 and the remaining values are -1, it is a suspected abnormal state and the next set of data is needed to determine whether abnormal characteristics appear;
[0063] If one or two of a, b, c, and d are 1 and the remaining values are 0, the network behavior is in a critical abnormal state, and a warning signal is output and the network behavior is continuously monitored; if three or more of a, b, c, and d are 1, the network behavior is in an abnormal state, and a warning signal is output and an alarm is promptly sent to the security management center.
[0064] If a, b, c, and d are all less than or equal to 1, the network behavior is normal; if one or two of a, b, c, and d are 1 and the remaining values are -1, it is a suspected abnormal state and the next set of data is needed to determine whether abnormal characteristics appear;
[0065] If one or two of a, b, c, and d are 1 and the remaining values are 0, the network behavior is in a critical abnormal state, and a warning signal is output and the network behavior is continuously monitored; if three or more of a, b, c, and d are 1, the network behavior is in an abnormal state, and a warning signal is output and an alarm is promptly sent to the security management center.
[0066] It should be noted that the core of the network behavior monitoring and positioning method is to use AI technology to identify and locate potential abnormal behaviors through comprehensive analysis of traffic data in the network. This process not only includes real-time monitoring of data, but also involves the classification and marking of abnormal behaviors, as well as rapid response to security threats.
[0067] Specifically, obtaining the traffic data of each node in the network involves the comprehensive capture of network traffic, which may include but is not limited to the size of the data packet, transmission rate, source address and destination address. By using AI analysis technology to analyze traffic data, historical data can be trained through machine learning algorithms to establish a behavior pattern library for real-time comparison and identification of abnormal patterns. The calculation of traffic feature quantities requires the definition of specific parameters, such as the observation period, the traffic intensity of the data packet, the height of the network layer, the transmission distance, and the average transmission efficiency of the network.
[0068] When implemented, the application of AI analysis technology in traffic data analysis is a comprehensive automated process involving multiple links such as data collection, feature extraction, model training, anomaly detection and adaptive learning. First, the system captures raw traffic data from the network, including the header information, payload content, transmission time, etc. of the data packet. Then, machine learning algorithms are used to automatically extract key features from these data, such as packet size, source and target IP addresses, port numbers, etc. These features can reveal traffic patterns and trends.
[0069] Furthermore, AI analysis technology learns and builds a behavioral pattern library of normal traffic by analyzing historical traffic data, which serves as a benchmark for judging whether the current traffic is abnormal. Subsequently, based on the extracted features and behavioral pattern library, the system trains the machine learning model to enable it to identify normal traffic and potential abnormal traffic. In real-time traffic analysis, the system compares the captured data with the behavioral pattern library to identify and mark possible abnormal patterns.
[0070] When the system detects traffic that does not conform to normal behavior patterns, it will be classified and marked as abnormal, and a detailed report will be generated based on the severity and type to provide decision support for analysts. In addition, AI analysis technology has adaptive learning capabilities and can continuously optimize the behavior pattern library and machine learning model based on new traffic data to adapt to the dynamic changes in the network environment. Ultimately, AI analysis technology can be integrated into the network security framework as an automated component to improve the speed and accuracy of response to complex network threats.
[0071] Preferably, the application of AI judgment technology can be further refined, for example, by selecting a specific machine learning model, such as a support vector machine (SVM) or a deep neural network (DNN), and adjusting the model parameters according to the actual network environment. At the same time, dynamic thresholds can be set and automatically adjusted according to the real-time status and historical data of the network to adapt to different network conditions and behavior patterns. In addition, a multi-factor decision-making system can be introduced to combine traffic feature quantities and other network indicators, such as user behavior analysis, device status, etc., to improve the accuracy and efficiency of abnormal behavior identification.
[0072] In some embodiments, edge computing nodes are used to perform real-time analysis when monitoring network traffic.
[0073] It should be noted that the concept of using edge computing nodes for real-time analysis when monitoring network traffic is based on the use of edge computing technology to disperse data processing and analysis tasks to the edge of the network, that is, close to the data source. This can reduce data transmission delays, improve analysis efficiency, and reduce dependence on central servers.
[0074] Specifically, edge computing nodes can be understood as small servers or dedicated devices deployed at the edge of the network. They have certain computing capabilities and can perform preliminary processing and analysis on the collected network traffic data. For example, these nodes can monitor the size of data packets, transmission rate, source and destination of data packets in real time. Parameter settings may include determining the frequency of data packet capture, the depth of analysis, and threshold settings, while concepts may include but are not limited to the type of data packet, protocol type, and traffic pattern.
[0075] Preferably, the configuration of edge computing nodes can be adjusted according to the specific needs and environment of the network. For example, in a high-traffic network environment, the number of edge computing nodes can be increased to improve processing power; in application scenarios with extremely high real-time requirements, algorithms can be optimized to reduce analysis delays. In addition, a variety of data preprocessing techniques, such as data cleaning and feature extraction, can be used to improve the accuracy of analysis. As an alternative, cloud computing resources can also be used to work in conjunction with edge computing nodes to achieve more efficient resource allocation and load balancing.
[0076] In some embodiments, the method for AI to identify malicious behavior patterns is to train historical traffic data through a machine learning algorithm to establish a behavior pattern library. When the degree of match between real-time traffic data and malicious behavior patterns in the library exceeds a preset threshold, it is identified as an abnormal state.
[0077] It should be noted that the AI method of identifying malicious behavior patterns is the core technical link. It trains historical traffic data through machine learning algorithms to establish a behavior pattern library. This library can store and identify a variety of network behavior patterns, including but not limited to normal communication patterns, known attack patterns, and potential unknown threat patterns.
[0078] Specifically, the machine learning algorithms involved in AI analysis technology can include supervised learning, unsupervised learning, or semi-supervised learning. These algorithms require a large amount of historical traffic data as a training set to identify and learn normal and abnormal network behavior characteristics. The parameter settings in the behavior pattern library, such as feature extraction methods, classifier types, and training algorithm parameters, all need to be adjusted and optimized according to the actual network environment and security requirements.
[0079] Preferably, it can be further refined into the following steps: first, select a suitable machine learning framework and algorithm, such as support vector machine (SVM), decision tree or deep neural network, etc.; second, define and extract key features in traffic data, such as the frequency, size, transmission time window of data packets, etc.; then, use these features to train the model, and evaluate the performance of the model through cross-validation and other techniques; finally, deploy the trained model to the network monitoring system, analyze the traffic data in real time and match it with the behavior pattern library. When the matching degree exceeds the preset threshold, the system will automatically identify and mark the abnormal behavior. In addition, incremental learning or online learning methods can be used to enable the model to adapt to the dynamic changes of network behavior and continuously improve the accuracy and adaptability of recognition.
[0080] In some embodiments, the preset threshold is a matching degree between 0.85 and 1.
[0081] It should be noted that the preset threshold is a key parameter for determining the matching degree between real-time traffic data and malicious behavior patterns in the behavior pattern library. The setting of this threshold directly affects the false alarm rate and missed alarm rate of the system, so it needs to be carefully selected to balance the two.
[0082] Specifically, the preset threshold is a value between 0.85 and 1, which is used to evaluate the similarity between traffic data and known malicious behavior patterns. For example, the threshold can be set to 0.9, which means that only when the real-time traffic data matches the pattern in the library at or above 90%, the system will identify it as an abnormal state. When setting parameters, it is necessary to consider the diversity and complexity of traffic data, as well as the coverage and update frequency of the malicious behavior pattern library.
[0083] Preferably, a method of dynamically adjusting the threshold can be used to adapt to different network environments and traffic characteristics. For example, when the network traffic is low or high, the system can automatically adjust the threshold to adapt to traffic changes and reduce false positives. In addition, the system can also automatically optimize the threshold setting based on the results of historical data analysis to improve recognition accuracy. As an alternative, a multi-threshold system can be introduced to set different matching thresholds according to different malicious behavior types or risk levels to achieve more refined abnormal behavior recognition.
[0084] In some embodiments, WiFi radio packets captured by a dedicated network card are positively correlated with abnormality of network behavior.
[0085] It should be noted that the WiFi radio messages captured by the dedicated network card are positively correlated with the abnormality of network behavior, which means that by analyzing the captured WiFi radio messages, abnormal behavior in the network can be discovered. This method uses the direct representation of network traffic data and provides an intuitive basis for the detection of abnormal behavior.
[0086] Specifically, a dedicated network card here refers to a hardware device designed specifically for capturing network traffic. It can monitor and capture data packets transmitted over a WiFi network in real time. These data packets may contain various information, such as IP addresses, port numbers, transmission protocols, etc., which are essential for analyzing network behavior. In terms of parameter settings, you can set the frequency of data packet capture, the size of captured data packets, and other parameters according to the network environment and security requirements to optimize the monitoring effect.
[0087] Preferably, in addition to the basic capture settings, the operation steps can be further refined. For example, specific data packet characteristics can be set as trigger conditions. When data packets that meet these characteristics are captured, the system will automatically perform a deeper analysis. In addition, machine learning algorithms can be used to classify and pattern recognize captured data packets to improve the accuracy of abnormal behavior detection. As an alternative, other types of network interface cards, such as Ethernet cards, can also be used to capture abnormal traffic in wired networks.
[0088] In some embodiments, the malicious probing behavior detected by the deception technology is positively correlated with the abnormality of the network behavior.
[0089] It should be noted that the malicious probing behavior detected by the deception technology is positively correlated with the abnormality of network behavior, which means that malicious activities in the network can be effectively identified through deception technology. Deception technology usually includes setting up trap networks or trap systems that simulate real network services to attract and detect attackers' probing behavior.
[0090] Specifically, the trap network in the deception technology can be a virtual server, database or other network resources, which are designed to easily attract the attention of attackers. These trap systems are usually deployed in key locations of the network to capture the attacker's detection behavior. In terms of parameter settings, the configuration of the trap network can be adjusted according to the network environment and the expected attack mode, such as open ports, service types, etc., to maximize the efficiency of trapping.
[0091] Preferably, automated deception technology deployment and management tools can be used, which can dynamically adjust the configuration of the trap network according to changes in network traffic and behavior patterns. For example, when a specific type of scanning behavior is detected, the system can automatically update the service configuration of the trap network to better adapt to and identify the attacker's detection strategy. In addition, the captured data can be analyzed in conjunction with machine learning algorithms to identify more complex malicious detection patterns, thereby improving the intelligence level and responsiveness of the entire monitoring system.
[0092] The above-mentioned embodiments of the present invention have the following beneficial effects: the network behavior monitoring and positioning method can significantly enhance the real-time monitoring and analysis capabilities of network traffic by integrating advanced AI judgment technology and edge computing capabilities. The method can quickly identify abnormal behavior and effectively classify it by intelligently analyzing traffic data, such as marking abnormal signals with traffic characteristics higher than the set threshold as blacklists, and marking signals with traffic characteristics equal to or less than the set threshold as critical state signals and normal state signals, respectively. In addition, the traffic characteristic calculation formula adopted by the method combines multiple key attributes of data packets, such as size, transmission rate, network layer height, transmission distance and average transmission efficiency, which can provide a more accurate quantitative evaluation of the abnormality of network behavior. Furthermore, the method can improve the recognition accuracy of malicious behavior patterns by training the behavior pattern library established by machine learning algorithms. The setting of the preset threshold ensures that only when the real-time traffic data matches the malicious behavior pattern in the library more than a specific value, it will be judged as an abnormal state. This method can not only improve the sensitivity and specificity of anomaly detection, but also further enhance the detection capability of malicious probing behavior through deception technology and packet capture analysis of dedicated network cards, thereby building a more comprehensive and efficient network security protection mechanism.
[0093] like Figure 2 As shown, a network behavior monitoring and positioning system 200 of some embodiments includes:
[0094] A data acquisition module is used to obtain the flow data of each node in the network;
[0095] The analysis module is used to analyze traffic data using AI analysis technology to determine whether there is abnormal behavior;
[0096] The marking module is used to mark and manage abnormal behaviors;
[0097] A judgment module is used to judge abnormal characteristics of network behavior;
[0098] A warning module is used to output a warning signal and perform corresponding processing when an abnormal state is detected;
[0099] The security management center is used to receive alarm signals and make further security responses.
[0100] It is understandable that the modules and references in the network behavior monitoring and positioning system 200 are Figure 1 Therefore, the operations, features and beneficial effects described above for the network behavior monitoring and positioning method are also applicable to the network behavior monitoring and positioning system 200 and the modules contained therein, and will not be repeated here.
[0101] Reference below Figure 3 , which shows a schematic diagram of a structure 300 of an electronic device suitable for implementing some embodiments of the present invention. The electronic devices in some embodiments of the present invention may include, but are not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.
[0102] like Figure 3 As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.
[0103] Typically, the following devices may be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3Each block shown in the figure may represent one device, or may represent multiple devices as required.
[0104] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.
[0105] The above descriptions are only some preferred embodiments of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present invention is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention to form a technical solution.
Claims
1. A network behavior monitoring and positioning method, characterized in that: include: Obtain traffic data of each node in the network; Use AI analysis technology to analyze traffic data and determine whether there is abnormal behavior; If it does not exist, the network behavior is normal. If it does exist, the abnormal behavior characteristics and their types are recorded; Among the abnormal behaviors analyzed, if the traffic characteristic quantity is higher than the set threshold, it is an abnormal signal and is marked as blacklisted; If the traffic characteristic is equal to the set threshold, it is a critical state signal and is marked as a whitelist; If the flow characteristic quantity is less than the set threshold, it is a normal state signal and is not marked; If AI identifies malicious behavior patterns and abnormal status signals, it will be marked as blacklisted; If AI does not identify malicious behavior patterns and it is a normal state signal, it will not be marked; If abnormal messages are found through packet capture and analysis of the dedicated network card of the edge computing node, which is an abnormal status signal, they will be marked as blacklisted; If no abnormal message is found, it is a normal state signal and no marking is performed; If the deception technology detects malicious probing behavior, which is a signal of abnormal status, it will be marked as blacklisted; If no malicious detection behavior is detected and it is a normal state signal, no marking will be performed; The flow characteristic calculation formula is: Among them, F is the traffic characteristic; n is the total number of data packets in the observation period; I i is the traffic intensity of the ith data packet, including the size or transmission rate of the data packet; h i is the network layer height of the ith data packet, including the number of network layers or routing hops that the data packet passes through; z i is the transmission distance of the ith data packet, including the network distance of the data packet from the source to the destination; σ is the average transmission efficiency of the network, including the inverse of the average data packet loss rate of the network; ω is the signal frequency of the network, including the average frequency of data packet transmission in the network; Based on the comparison results of the traffic feature F and the set threshold, as well as the results of AI analysis, packet capture analysis, and deception technology detection, the abnormal indicators a, b, c, and d of the network behavior are determined, where: If F is higher than the set threshold, a=1; if it is equal to, a=0; if it is lower, a=-1; If AI identifies a malicious behavior pattern with a matching degree exceeding the preset threshold, b = 1; if not, b = 0; If an abnormal message is found through packet capture analysis by the dedicated network card, c = 1; if no abnormal message is found, c = 0; If the deception technology detects malicious probing behavior, then d = 1; if not, then d = 0; If a, b, c, and d are all less than or equal to 1, the network behavior is normal; if one or two of a, b, c, and d are 1 and the remaining values are -1, it is a suspected abnormal state and the next set of data is needed to determine whether abnormal characteristics appear; If one or two of a, b, c, and d are 1 and the remaining values are 0, the network behavior is in a critical abnormal state, and a warning signal is output and the network behavior is continuously monitored; if three or more of a, b, c, and d are 1, the network behavior is in an abnormal state, and a warning signal is output and an alarm is promptly sent to the security management center.
2. A network behavior monitoring and positioning method according to claim 1, characterized in that: When monitoring network traffic, edge computing nodes are used for real-time analysis.
3. A network behavior monitoring and positioning method according to claim 1, characterized in that: The method used by AI to identify malicious behavior patterns is to train historical traffic data through machine learning algorithms and establish a behavior pattern library. When the degree of match between real-time traffic data and malicious behavior patterns in the library exceeds a preset threshold, it is identified as an abnormal state.
4. A network behavior monitoring and positioning method according to claim 3, characterized in that: The preset threshold is a matching degree between 0.85 and 1.
5. A network behavior monitoring and positioning method according to claim 1, characterized in that: WiFi radio messages captured by dedicated network cards are positively correlated with the abnormality of network behavior.
6. A network behavior monitoring and positioning method according to claim 1, characterized in that: The malicious probing behavior detected by deception technology is positively correlated with the abnormality of network behavior.
7. A network behavior monitoring and positioning system, characterized in that: include: A data acquisition module is used to obtain the flow data of each node in the network; The analysis module is used to analyze traffic data using AI analysis technology to determine whether there is abnormal behavior; The marking module is used to mark and manage abnormal behaviors; A judgment module is used to judge abnormal characteristics of network behavior; A warning module is used to output a warning signal and perform corresponding processing when an abnormal state is detected; The security management center is used to receive alarm signals and make further security responses.
