Software security test system based on function coverage and vulnerability scanning

Abstract

The invention discloses a software security test system based on function coverage and vulnerability scanning. Comprising a test management module, a function coverage module, a vulnerability scanning module, a vulnerability repairing and updating module, an artificial intelligence analysis module, a report generation module, a security feedback loop module, an enhanced dynamic adaptation module and a security knowledge graph module. The test management module is used for being responsible for distribution, monitoring and management of test tasks and supporting multi-thread and distributed test; and the function coverage module comprises a static code analysis unit and a dynamic operation analysis unit, and is used for determining function points of the software through static code analysis and dynamic operation analysis, generating corresponding test cases and ensuring that each function point is covered. According to the method, function coverage, vulnerability scanning and artificial intelligence analysis are combined, so that the function correctness and safety of test software are improved, potential risks can be predicted, multi-thread and distributed testing is adopted, and the test efficiency is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of software testing, and in particular relates to a software security testing system based on function coverage and vulnerability scanning. Background Art

[0002] With the rapid development of information technology, software has become an indispensable part of daily life and work. However, the software security issues that come with it have become increasingly prominent, posing significant risks to individuals, enterprises and even countries. Software security testing is a method of evaluating and testing the security of a system, application or network. It aims to discover vulnerabilities and weaknesses in the system, as well as evaluate the system's ability to resist various security threats. Through security testing, organizations can identify security vulnerabilities and threats in the system, and provide improvement suggestions and solutions to improve the organization's security and protect its information resources.

[0003] Traditional software testing methods often focus on functional testing and ignore security testing, which leads to many potential security risks not being discovered and fixed in time. Even if security vulnerabilities are found, how to effectively fix these vulnerabilities and verify the effectiveness of the repair measures is also a major challenge. In addition, security testing requires testers to have deep security knowledge and rich practical experience. However, professional talents in this area are relatively scarce, resulting in many organizations facing a shortage of manpower when conducting security testing. To this end, we propose a software security testing system based on functional coverage and vulnerability scanning. Summary of the invention

[0004] The purpose of the present invention is to provide a software security testing system based on function coverage and vulnerability scanning to solve the problems raised in the above background technology.

[0005] To achieve the above-mentioned object, the present invention provides the following technical solutions: a software security testing system based on function coverage and vulnerability scanning, comprising a test management module, a function coverage module, a vulnerability scanning module, a vulnerability patching and updating module, an artificial intelligence analysis module, a report generation module, a security feedback loop module, an enhanced dynamic adaptation module and a security knowledge graph module;

[0006] The test management module is responsible for the allocation, monitoring and management of test tasks, and supports multi-threaded and distributed testing;

[0007] The functional coverage module includes a static code analysis unit and a dynamic operation analysis unit, which are used to determine the functional points of the software through static code analysis and dynamic operation analysis, and generate corresponding test cases to ensure that each functional point is covered;

[0008] The vulnerability scanning module is used to integrate multiple vulnerability scanning tools to perform in-depth scanning on the software and discover potential security vulnerabilities;

[0009] The vulnerability patching and updating module is used to automatically or manually patch potential security vulnerabilities when they are discovered, and to update the security vulnerability database in a timely manner to ensure that the software remains up to date and in the most secure state;

[0010] The artificial intelligence analysis module is used to use machine learning algorithms to perform intelligent analysis on test results, predict unknown vulnerabilities, and optimize test strategies;

[0011] The security feedback loop module is used to collect user feedback and real-time threat intelligence, dynamically adjust the test strategy and vulnerability database, and ensure that the system continues to improve the test quality over time;

[0012] The report generation module is used to generate a detailed security test report based on the results of function coverage, vulnerability scanning, artificial intelligence analysis and security feedback loop, and the security test report includes vulnerability description, risk level and repair suggestions;

[0013] The enhanced dynamic adaptation module is used to perceive and understand the state of the current test environment, dynamically generate or adjust the test script based on the perceived environment state data, and provide real-time, adaptive test environment configuration and optimization;

[0014] The security knowledge graph module is used to build a comprehensive knowledge graph that brings together various security vulnerabilities, attack patterns, defense strategies and related cases to improve the intelligence level and testing quality of the system.

[0015] Preferably, the static code analysis unit is used to discover information about program structure, dependencies, and control flow by scanning source code or compiled binary code without actually executing the program, and to build a logical model of the software using abstract syntax trees and control flow graph technology to identify all functions, classes, variables, and the calling relationships between them, ensuring theoretical coverage of all executable paths.

[0016] Preferably, the dynamic operation analysis unit is used to monitor the execution process of the software while it is running, record the function call sequence, the information transmitted by the data flow, and implant monitoring code at key code points through the plug-in technology to collect runtime data, so as to supplement the deficiencies of static analysis with the actual execution path and achieve a higher level of coverage.

[0017] Preferably, the vulnerability scanning module includes a feature matching unit, a behavior analysis unit and a machine learning auxiliary unit. The feature matching unit performs pattern matching on software code or runtime behavior based on a feature database of known vulnerabilities to quickly identify known security vulnerabilities; the behavior analysis unit is used to deeply analyze the behavioral patterns of software runtime, such as input verification, output filtering, and permission control, to identify unsafe operations or logical defects that cause security issues; the machine learning auxiliary unit is used to introduce machine learning algorithms, train historical vulnerability data, learn vulnerability features, and enhance the ability to identify new or variant vulnerabilities.

[0018] Preferably, the enhanced dynamic adaptation module includes an environment perception unit, a test script generation unit, a decision optimization unit and a troubleshooting unit. The environment perception unit is used to perceive and understand the status of the current test environment, including the operating system, hardware configuration, and network status, and automatically adjust the test parameters based on this information; the test script generation unit is used to dynamically generate or adjust the test script based on the perceived environmental status data to maximize the test coverage and efficiency; the decision optimization unit is used to make optimization decisions in real time through a reinforcement learning algorithm; and the troubleshooting unit is used to automatically diagnose the cause of the problem when an unexpected error or failure is encountered during the test, and provide a solution or automatic repair.

[0019] Preferably, the security knowledge graph module includes a knowledge integration unit, a graph construction unit, a recommendation unit and a continuous learning unit. The knowledge integration unit is used to integrate security-related knowledge from different sources to form a unified knowledge base, and the different sources include publicly released security vulnerability databases, the latest security research papers and known attack cases; the graph construction unit is used to construct a knowledge graph using graph database technology to structure various security concepts and entities and the relationships between them; the recommendation unit is used to recommend the most suitable test strategy, potential risk points and possible attack paths according to the specific situation of the current test object during the testing process; the continuous learning unit is used to continuously learn and evolve from new data, update the content of the knowledge graph, and keep synchronized with the latest security threats.

[0020] Preferably, the intelligent analysis of the test results by using a machine learning algorithm is specifically performed by using a logistic regression algorithm, and the logistic regression algorithm includes the following formula:

[0021]

[0022] Among them, P represents probability, that is, the probability of output variable y=1 given the input feature variable x. In logistic regression, this probability is always between 0 and 1; w and b are model parameters, and x is the input feature vector containing all the predictor variables.

[0023] Preferably, the function coverage module includes coverage calculation, and the formula for coverage calculation includes:

[0024] Coverage Percentage=(Covered Points / Total Points)x100%;

[0025] Among them, Coverage Percentage represents the basic coverage, Covered Points represents the number of points covered by the test, and Total Points represents the total number of points that need to be covered.

[0026] Preferably, the vulnerability scanning module evaluates the severity of the vulnerability through a CVSS scoring system, wherein the score in the CVSS scoring system is composed of three different metric categories, including a base metric, a time metric, and an environmental metric.

[0027] Preferably, the basic metric comprises the following formula:

[0028] Calculate the base impact sub-score:

[0029]

[0030] Among them, C represents the maximum value among Confidentiality Impact (CI), Integrity Impact (II), and Availability Impact (AI), and I represents the average of Confidentiality Impact, Integrity Impact, and Availability Impact;

[0031] Calculate the basic Exploitability = 8.22 × AV × AC × UI;

[0032] AV stands for attack vector, which measures the path or distance an attacker can take to exploit a vulnerability; AC stands for attack complexity, which describes the complexity of the skills and / or preparation required to exploit a vulnerability; UI stands for user interaction, which refers to whether the target system user needs to participate in successfully exploiting the vulnerability;

[0033] Calculate the base score:

[0034] Base_Score=min(1.08×(Exploitability+Impact_Adj),10);

[0035] Impact_Adj is a value adjusted based on S and I_Score, and S represents the scope of impact after the vulnerability is exploited.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] (1) The present invention combines functional coverage, vulnerability scanning and artificial intelligence analysis to not only improve the functional correctness and security of the test software, but also predict potential risks, and adopts multi-threaded and distributed testing to improve testing efficiency;

[0038] (2) The present invention can further optimize the test process by setting up an artificial intelligence analysis module, and can ensure that the system continues to improve the test quality over time by setting up a security feedback loop module. In addition, in conjunction with the report generation module, it can automatically generate test cases and security test reports, thereby improving the comprehensiveness and accuracy of the test;

[0039] (3) The present invention can identify and test more security vulnerabilities by setting up an artificial intelligence analysis module, an enhanced dynamic adaptation module and a security knowledge graph module, using machine learning algorithms and security knowledge graphs, thereby ensuring the comprehensiveness of the test. By combining the integrated security knowledge graph and the enhanced dynamic adaptation module, the security test can be automatically performed, reducing manual intervention, which is conducive to solving the problem of insufficient manpower, thereby greatly improving the efficiency of the test;

[0040] (4) The vulnerability scanning module and vulnerability patching and updating module provided in the present invention can timely discover and repair security vulnerabilities, effectively reducing the security threats and potential losses faced by the system after deployment. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a structural block diagram of the present invention;

[0042] Figure 2 It is a structural block diagram of the vulnerability scanning module of the present invention;

[0043] Figure 3 It is a structural block diagram of the enhanced dynamic adaptation module of the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] See also Figure 1-Figure 3, the present invention provides a technical solution: a software security testing system based on function coverage and vulnerability scanning, including a test management module, a function coverage module, a vulnerability scanning module, a vulnerability patching and updating module, an artificial intelligence analysis module, a report generation module, a security feedback loop module, an enhanced dynamic adaptation module and a security knowledge graph module;

[0046] The test management module is responsible for the allocation, monitoring and management of test tasks, and supports multi-threaded and distributed testing;

[0047] The functional coverage module includes a static code analysis unit and a dynamic operation analysis unit, which are used to determine the functional points of the software through static code analysis and dynamic operation analysis, and generate corresponding test cases to ensure that each functional point is covered;

[0048] The vulnerability scanning module is used to integrate multiple vulnerability scanning tools to perform in-depth scanning on the software and discover potential security vulnerabilities;

[0049] The vulnerability patching and updating module is used to automatically or manually patch potential security vulnerabilities when they are discovered, and to update the security vulnerability database in a timely manner to ensure that the software remains up to date and in the most secure state;

[0050] The artificial intelligence analysis module is used to use machine learning algorithms to perform intelligent analysis on test results, predict unknown vulnerabilities, and optimize test strategies;

[0051] The security feedback loop module is used to collect user feedback and real-time threat intelligence, dynamically adjust the test strategy and vulnerability database, and ensure that the system continues to improve the test quality over time;

[0052] The report generation module is used to generate a detailed security test report based on the results of function coverage, vulnerability scanning, artificial intelligence analysis and security feedback loop, and the security test report includes vulnerability description, risk level and repair suggestions;

[0053] The enhanced dynamic adaptation module is used to perceive and understand the state of the current test environment, dynamically generate or adjust the test script based on the perceived environment state data, and provide real-time, adaptive test environment configuration and optimization;

[0054] The security knowledge graph module is used to build a comprehensive knowledge graph that brings together various security vulnerabilities, attack patterns, defense strategies and related cases to improve the intelligence level and testing quality of the system.

[0055] In this embodiment, preferably, the static code analysis unit is used to discover information about program structure, dependencies, and control flow by scanning source code or compiled binary code without actually executing the program, and uses abstract syntax trees and control flow graph technology to build a logical model of the software to identify all functions, classes, variables, and the calling relationships between them, ensuring theoretical coverage of all executable paths.

[0056] In this embodiment, preferably, the dynamic operation analysis unit is used to monitor the execution process of the software while it is running, record the function call sequence, the information transmitted by the data flow, and implant monitoring code at key code points through the insertion technology to collect runtime data, so as to supplement the deficiencies of static analysis with the actual execution path and achieve a higher level of coverage.

[0057] In this embodiment, preferably, the vulnerability scanning module includes a feature matching unit, a behavior analysis unit and a machine learning auxiliary unit. The feature matching unit performs pattern matching on software code or runtime behavior based on a feature database of known vulnerabilities to quickly identify known security vulnerabilities; the behavior analysis unit is used to deeply analyze the behavioral patterns of software runtime, such as input verification, output filtering, and permission control, to identify unsafe operations or logical defects that cause security issues; the machine learning auxiliary unit is used to introduce machine learning algorithms, train historical vulnerability data, learn vulnerability features, and enhance the ability to identify new or variant vulnerabilities.

[0058] In this embodiment, preferably, the enhanced dynamic adaptation module includes an environment perception unit, a test script generation unit, a decision optimization unit and a troubleshooting unit. The environment perception unit is used to perceive and understand the status of the current test environment, including the operating system, hardware configuration, and network status, and automatically adjust the test parameters based on this information; the test script generation unit is used to dynamically generate or adjust the test script based on the perceived environmental status data to maximize the test coverage and efficiency; the decision optimization unit is used to make optimization decisions in real time through a reinforcement learning algorithm; the troubleshooting unit is used to automatically diagnose the cause of the problem when an unexpected error or failure is encountered during the test, and provide a solution or automatic repair.

[0059] In this embodiment, preferably, the security knowledge graph module includes a knowledge integration unit, a graph construction unit, a recommendation unit and a continuous learning unit. The knowledge integration unit is used to integrate security-related knowledge from different sources to form a unified knowledge base, and the different sources include publicly released security vulnerability databases, the latest security research papers and known attack cases; the graph construction unit is used to construct a knowledge graph using graph database technology to structure various security concepts and entities and the relationships between them; the recommendation unit is used to recommend the most suitable test strategy, potential risk points and possible attack paths according to the specific situation of the current test object during the testing process; the continuous learning unit is used to continuously learn and evolve from new data, update the content of the knowledge graph, and keep synchronized with the latest security threats.

[0060] In this embodiment, preferably, the intelligent analysis of the test results by using a machine learning algorithm is specifically performed by using a logistic regression algorithm, and the logistic regression algorithm includes the following formula:

[0061]

[0062] Among them, P represents probability, that is, the probability of output variable y=1 given the input feature variable x. In logistic regression, this probability is always between 0 and 1; w and b are model parameters, and x is the input feature vector containing all the predictor variables.

[0063] In this embodiment, preferably, the function coverage module includes coverage calculation, and the formula for coverage calculation includes:

[0064] Coverage Percentage=(Covered Points / Total Points)x100%;

[0065] Among them, Coverage Percentage represents the basic coverage, Covered Points represents the number of points covered by the test, and Total Points represents the total number of points that need to be covered.

[0066] In this embodiment, preferably, the vulnerability scanning module evaluates the severity of the vulnerability through a CVSS scoring system, in which the score consists of three different metric categories, including basic metrics, time metrics, and environmental metrics.

[0067] In this embodiment, preferably, the basic metric includes the following formula:

[0068] Calculate the base impact sub-score:

[0069]

[0070] Among them, C represents the maximum value among Confidentiality Impact (CI), Integrity Impact (II), and Availability Impact (AI), and I represents the average of Confidentiality Impact, Integrity Impact, and Availability Impact;

[0071] Calculate the basic Exploitability = 8.22 × AV × AC × UI;

[0072] AV stands for attack vector, which measures the path or distance an attacker can take to exploit a vulnerability; AC stands for attack complexity, which describes the complexity of the skills and / or preparation required to exploit a vulnerability; UI stands for user interaction, which refers to whether the target system user needs to participate in successfully exploiting the vulnerability;

[0073] Calculate the base score:

[0074] Base_Score=min(1.08×(Exploitability+Impact_Adj),10);

[0075] Impact_Adj is a value adjusted based on S and I_Score, and S represents the scope of impact after the vulnerability is exploited.

[0076] Principles and advantages of the present invention:

[0077] By combining function coverage, vulnerability scanning and artificial intelligence analysis, the present invention not only increases the functional correctness and security of the test software, but also predicts potential risks, and adopts multi-threaded and distributed testing to improve the test efficiency; by setting the artificial intelligence analysis module, the test process can be further optimized, and by setting the security feedback loop module, it can ensure that the system continues to improve the test quality over time, and in conjunction with the report generation module, it can automatically generate test cases and security test reports to improve the comprehensiveness and accuracy of the test; by setting the artificial intelligence analysis module, the enhanced dynamic adaptation module and the security knowledge graph module, using machine learning algorithms and the security knowledge graph, it can identify and test more security vulnerabilities, ensuring the comprehensiveness of the test, and by combining the integrated security knowledge graph and the enhanced dynamic adaptation module, it can automatically perform security testing, reduce manual intervention, and help solve the problem of insufficient manpower, thereby greatly improving the efficiency of the test; by setting the vulnerability scanning module and the vulnerability patching and updating module, security vulnerabilities can be discovered and repaired in a timely manner, effectively reducing the security threats and potential losses faced by the system after deployment.

[0078] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A software security testing system based on function coverage and vulnerability scanning, characterized by: It includes test management module, function coverage module, vulnerability scanning module, vulnerability patching and updating module, artificial intelligence analysis module, report generation module, security feedback loop module, enhanced dynamic adaptation module and security knowledge graph module; The test management module is responsible for the allocation, monitoring and management of test tasks, and supports multi-threaded and distributed testing; The functional coverage module includes a static code analysis unit and a dynamic operation analysis unit, which are used to determine the functional points of the software through static code analysis and dynamic operation analysis, and generate corresponding test cases to ensure that each functional point is covered; The vulnerability scanning module is used to integrate multiple vulnerability scanning tools to perform in-depth scanning on the software and discover potential security vulnerabilities; The vulnerability patching and updating module is used to automatically or manually patch potential security vulnerabilities when they are discovered, and to update the security vulnerability database in a timely manner to ensure that the software remains up to date and in the most secure state; The artificial intelligence analysis module is used to use machine learning algorithms to perform intelligent analysis on test results, predict unknown vulnerabilities, and optimize test strategies; The security feedback loop module is used to collect user feedback and real-time threat intelligence, dynamically adjust the test strategy and vulnerability database, and ensure that the system continues to improve the test quality over time; The report generation module is used to generate a detailed security test report based on the results of function coverage, vulnerability scanning, artificial intelligence analysis and security feedback loop, and the security test report includes vulnerability description, risk level and repair suggestions; The enhanced dynamic adaptation module is used to perceive and understand the state of the current test environment, dynamically generate or adjust the test script based on the perceived environment state data, and provide real-time, adaptive test environment configuration and optimization; The security knowledge graph module is used to build a comprehensive knowledge graph that brings together various security vulnerabilities, attack patterns, defense strategies and related cases to improve the intelligence level and testing quality of the system.

2. A software security testing system based on function coverage and vulnerability scanning according to claim 1, characterized in that: The static code analysis unit is used to discover information about program structure, dependencies, and control flow by scanning source code or compiled binary code without actually executing the program, and to build a logical model of the software using abstract syntax trees and control flow graph technology to identify all functions, classes, variables, and the calling relationships between them, ensuring theoretical coverage of all executable paths.

3. A software security testing system based on function coverage and vulnerability scanning according to claim 2, characterized in that: The dynamic operation analysis unit is used to monitor the execution process of the software while it is running, record the function call sequence and the information transmitted by the data flow, and implant monitoring code at key code points through the insertion technology to collect runtime data, so as to supplement the deficiencies of static analysis with the actual execution path and achieve a higher level of coverage.

4. The software security testing system based on function coverage and vulnerability scanning according to claim 1, characterized in that: The vulnerability scanning module includes a feature matching unit, a behavior analysis unit and a machine learning auxiliary unit. The feature matching unit performs pattern matching on software code or runtime behavior based on a feature database of known vulnerabilities to quickly identify known security vulnerabilities; the behavior analysis unit is used to deeply analyze the behavioral patterns of software runtime, such as input verification, output filtering, and permission control, to identify unsafe operations or logical defects that cause security issues; the machine learning auxiliary unit is used to introduce machine learning algorithms, train historical vulnerability data, learn vulnerability features, and enhance the ability to identify new or variant vulnerabilities.

5. The software security testing system based on function coverage and vulnerability scanning according to claim 1, characterized in that: The enhanced dynamic adaptation module includes an environment perception unit, a test script generation unit, a decision optimization unit and a troubleshooting unit. The environment perception unit is used to perceive and understand the status of the current test environment, including the operating system, hardware configuration, and network status, and automatically adjust the test parameters according to this information; the test script generation unit is used to dynamically generate or adjust the test script based on the perceived environmental status data to maximize the test coverage and efficiency; The decision optimization unit is used to make optimization decisions in real time through a reinforcement learning algorithm; the troubleshooting unit is used to automatically diagnose the cause of the problem when an unexpected error or failure is encountered during testing, and to provide a solution or automatic repair.

6. A software security testing system based on function coverage and vulnerability scanning according to claim 1, characterized in that: The security knowledge graph module includes a knowledge integration unit, a graph construction unit, a recommendation unit and a continuous learning unit. The knowledge integration unit is used to integrate security-related knowledge from different sources to form a unified knowledge base. The different sources include publicly released security vulnerability databases, the latest security research papers and known attack cases; the graph construction unit is used to construct a knowledge graph using graph database technology to structure various security concepts and entities and the relationships between them; the recommendation unit is used to recommend the most suitable test strategy, potential risk points and possible attack paths according to the specific situation of the current test object during the testing process; the continuous learning unit is used to continuously learn and evolve from new data, update the content of the knowledge graph, and keep pace with the latest security threats.

7. The software security testing system based on function coverage and vulnerability scanning according to claim 1, characterized in that: The intelligent analysis of the test results by using a machine learning algorithm is specifically performed by using a logistic regression algorithm, and the logistic regression algorithm includes the following formula: Among them, P represents probability, that is, the probability of output variable y=1 given the input feature variable x. In logistic regression, this probability is always between 0 and 1; w and b are model parameters, and x is the input feature vector containing all the predictor variables.

8. The software security testing system based on function coverage and vulnerability scanning according to claim 1, characterized in that: The functional coverage module includes coverage calculation, and the formula for coverage calculation includes: Coverage Percentage=(Covered Points / Total Points)x100%; Among them, Coverage Percentage represents the basic coverage, Covered Points represents the number of points covered by the test, and Total Points represents the total number of points that need to be covered.

9. The software security testing system based on function coverage and vulnerability scanning according to claim 1, characterized in that: The vulnerability scanning module evaluates the severity of the vulnerability through the CVSS scoring system, in which the score consists of three different metric categories, including basic metrics, time metrics, and environmental metrics.

10. A software security testing system based on function coverage and vulnerability scanning according to claim 9, characterized in that: The basic metric includes the following formula: Calculate the base impact sub-score: Among them, C represents the maximum value among ConfidentialityImpact (CI), IntegrityImpact (II), and AvailabilityImpact (AI), and I represents the average of Confidentiality Impact, Integrity Impact, and Availability Impact; Calculate the basic Exploitability = 8.22 × AV × AC × UI; AV stands for attack vector, which measures the path or distance an attacker can take to exploit a vulnerability; AC stands for attack complexity, which describes the complexity of the skills and / or preparation required to exploit a vulnerability; UI stands for user interaction, which refers to whether the target system user needs to participate in successfully exploiting the vulnerability; Calculate the base score: Base_Score=min(1.08×(Exploitability+Impact_Adj),10); Impact_Adj is a value adjusted based on S and I_Score, and S represents the scope of impact after the vulnerability is exploited.

Images