Vulnerability Detection Methods Research Articles

Open source software security vulnerability detection based on dynamic behavior features.

Open source software has been widely used in various industries due to its openness and flexibility, but it also brings potential security problems. Therefore, security analysis is required before using open source software. The current mainstream open source software vulnerability analysis technology is based on source code, and there are problems such as false positives, false negatives and restatements. In order to solve the problems, based on the further study of behavior feature extraction and vulnerability detection technology, a method of using dynamic behavior features to detect open source software vulnerabilities is proposed. Firstly, the relationship between open source software vulnerability and API call sequence is studied. Then, the behavioral risk vulnerability database of open source software is proposed as a support for vulnerability detection. In addition, the CNN-IndRNN classification model is constructed by improving the Independently Recurrent Neural Net-work (IndRNN) algorithm and applies to open source software security vulnerability detection. The experimental results verify the effectiveness of the proposed open source software security vulnerability detection method based on dynamic behavior features.

μVulDeePecker: A Deep Learning-Based System for Multiclass Vulnerability Detection

Fine-grained software vulnerability detection is an important and challenging problem. Ideally, a detection system (or detector) not only should be able to detect whether or not a program contains vulnerabilities, but also should be able to pinpoint the type of a vulnerability in question. Existing vulnerability detection methods based on deep learning can detect the presence of vulnerabilities (i.e., addressing the binary classification or detection problem), but cannot pinpoint types of vulnerabilities (i.e., incapable of addressing multiclass classification). In this paper, we propose the first deep learning-based system for multiclass vulnerability detection, dubbed μ VulDeePecker. The key insight underlying μ VulDeePecker is the concept of code attention, which can capture information that can help pinpoint types of vulnerabilities, even when the samples are small. For this purpose, we create a dataset from scratch and use it to evaluate the effectiveness of μ VulDeePecker. Experimental results show that μ VulDeePecker is effective for multiclass vulnerability detection and that accommodating control-dependence (other than data-dependence) can lead to higher detection capabilities.

An Automated Vulnerability Detection and Remediation Method for Software Security

As hacking techniques become more sophisticated, vulnerabilities have been gradually increasing. Between 2010 and 2015, around 80,000 vulnerabilities were newly registered in the CVE (Common Vulnerability Enumeration), and the number of vulnerabilities has continued to rise. While the number of vulnerabilities is increasing rapidly, the response to them relies on manual analysis, resulting in a slow response speed. It is necessary to develop techniques that can detect and patch vulnerabilities automatically. This paper introduces a trend of techniques and tools related to automated vulnerability detection and remediation. We propose an automated vulnerability detection method based on binary complexity analysis to prevent a zero-day attack. We also introduce an automatic patch generation method through PLT/GOT table modification to respond to zero-day vulnerabilities.

Statistics of software vulnerability detection in certification testing

The paper discusses practical aspects of introduction of the methods to detect software vulnerability in the day-to-day activities of the accredited testing laboratory. It presents the approval results of the vulnerability detection methods as part of the study of the open source software and the software that is a test object of the certification tests under information security requirements, including software for communication networks. Results of the study showing the allocation of identified vulnerabilities by types of attacks, country of origin, programming languages used in the development, methods for detecting vulnerability, etc. are given. The experience of foreign information security certification systems related to the detection of certified software vulnerabilities is analyzed. The main conclusion based on the study is the need to implement practices for developing secure software in the development life cycle processes. The conclusions and recommendations for the testing laboratories on the implementation of the vulnerability analysis methods are laid down.

An integration testing framework and evaluation metric for vulnerability mining methods

Software vulnerability mining is an important way to detect whether there are some loopholes existing in the software, and also is an important way to ensure the security of information systems. With the rapid development of information technology and software industry, most of the software has not been rigorously tested before being put in use, so that the hidden vulnerabilities in software will be exploited by the attackers. Therefore, it is of great significance for us to actively detect the software vulnerabilities in the security maintenance of information systems. In this paper, we firstly studied some of the commonly used vulnerability detection methods and detection tools, and analyzed the advantages and disadvantages of each method in different scenarios. Secondly, we designed a set of evaluation criteria for different mining methods in the loopholes evaluation. Thirdly, we also proposed and designed an integration testing framework, on which we can test the typical static analysis methods and dynamic mining methods as well as make the comparison, so that we can obtain an intuitive comparative analysis for the experimental results. Finally, we reported the experimental analysis to verify the feasibility and effectiveness of the proposed evaluation method and the testing framework, with the results showing that the final test results will serve as a form of guidance to aid the selection of the most appropriate and effective method or tools in vulnerability detection activity.

Research on Vulnerability Detection Technology for WEB Mail System

Recently, the Email system is seriously threatened by the vulnerability attack, and XSS vulnerability is one of the most serious vulnerability of WEB mail system. In this paper, we proposed a crossing site script injection vulnerability detection method for WEB mail system, and developed a detection tool for a systematic. With the method we proposed, we can automatically construct attack vectors according to different attack surfaces, and generate malformed mails for vulnerability detection. By applying the tool to some popular Webmail systems, seven exploitable XSS vulnerabilities are found. The evaluation result shows that our tool can efficiently detect XSS vulnerabilities.

Evaluation of Software Vulnerability Detection Methods and Tools: A Review

Evaluation of Software Vulnerability Detection Methods and Tools: A Review

Towards designing an extendable vulnerability detection method for executable codes

Context: Software vulnerabilities allow the attackers to harm the computer systems. Timely detection and removal of vulnerabilities help to improve the security of computer systems and avoid the losses from exploiting the vulnerabilities.Objective: Various methods have been proposed to detect the vulnerabilities in the past decades. However, most of these methods are suggested for detecting one or a limited number of vulnerability classes and require fundamental changes to be able to detect other vulnerabilities.In this paper, we present a first step towards designing an extendable vulnerability detection method that is independent from the characteristics of specific vulnerabilities.Method: To do so, we first propose a general model for specifying software vulnerabilities. Based on this model, a general specification method and an extendable algorithm is then presented for detecting the specified vulnerabilities in executable codes.As the first step, single-instruction vulnerabilities–the vulnerabilities that appear in one instruction–are specified and detected. We present a formal definition for single-instruction vulnerabilities. In our method, detection of the specified vulnerabilities is considered as solving a satisfaction problem. The suggested method is implemented as a plug-in for Valgrind binary instrumentation framework and the vulnerabilities are specified by the use of Valgrind intermediate language, called Vex.Results: Three classes of single-instruction vulnerabilities are specified in this paper, i. e. division by zero, integer bugs and NULL pointer dereference. The experiments demonstrate that the presented specification for these vulnerabilities are accurate and the implemented method can detect all the specified vulnerabilities.Conclusion:As we employ a general model for specifying the vulnerabilities and the structure of our vulnerability detection method does not depend on a specific vulnerability, our method can be extended to detect other specified vulnerabilities.

基于爬虫技术的Web应用程序漏洞检测方法

随着Web应用不断的发展，随之而产生的包括XSS在内的各种安全漏洞也越来越多。今天，XSS传统防御技术的缺陷已经越来越多地显现，例如防御种类单一、防御强度低、防御手段落后等，这就迫切需要不断提高和完善防御的方法和手段。针对此问题，提出了一种基于Scrapy的爬虫框架的Web应用程序漏洞检测方法。通过框架提供的便利条件对页面进行提取分析，根据不同的攻击方式生成特有的攻击向量，最后使页面注入点与攻击向量组合达到测试是否具有漏洞的目的。实验结果表明，这种漏洞检测方法在爬取页面以及漏洞检测的效率上都有了很大的提高。 With the continuous development of Web applications, a variety of security vulnerabilities, in-cluding XSS, also generate more and more. Today, the defects of the traditional XSS defense tech-nology have been more and more appear, such as a single type of defense, defense strength low, defense means backward. There is an urgent need to continuously improve and perfect the me-thods and means of defense. Aiming at this problem, this paper proposes a Web application vul-nerability detection method based on Scrapy. Through the framework to provide convenient con-ditions to the page for extraction and analysis, specific attack vector is generated according to the different ways of attacks. Finally, we make the combination of page injection points and attack vector to achieve the objective to test whether it is vulnerable. Experimental results show that this vulnerability detection method has a great improvement in the efficiency of crawling pages and vulnerability detection.

Software Vulnerability Detection Methodology Combined with Static and Dynamic Analysis

Software vulnerability is the attack surface. Therefore, vulnerabilities innate in software should be detected for software security assurance. Vulnerability detection method can be divided into static vulnerability detection and dynamic vulnerability detection. Static vulnerability detection is more commonly used for vulnerability detection. This method has many benefits, but it also creates false positives. Therefore, this paper proposes a method to combine static and dynamic detection to reduce false positives created from static vulnerability detection. The proposed method verifies the vulnerability by implanting a fault, based on the information received from static code analysis.

Research on Vulnerability Detection for Software Based on Taint Analysis

At present, Cross Site Scripting (XSS) vulnerability exists in most web sites. The main reason is the lack of effective validation and filtering mechanisms for user input data from web request. This paper explores vulnerability detection method which based on taint dependence analysis and implements a prototype system for Java Web program. We treat all user input as tainted data, and track the flow of Web applications, then we judge whether it will trigger an attack or not. The taint dependent analysis algorithm mentioned in this paper is used to construct the taint dependency graph. Next the value representation method of the string tainted object based on finite state automata is discussed. Finally, we propose the vulnerability detection method for the program. The experiment result shows that the prototype system can detect reflection cross-site scripting vulnerability well in those programs which dont have effective treatment for the user input data.

Defeating SQL Injection

The best strategy for combating SQL injection, which has emerged as the most widespread website security risk, calls for integrating defensive coding practices with both vulnerability detection and runtime attack prevention methods.

A novel vulnerability detection method for ZigBee MAC layer

Due to the hostile environment, open communication and implementation vulnerability, ZigBee is vulnerable to various attacks. But there are few effective vulnerability detection methods. In this paper, we design the ZigBee Border Conditions Based Tester ZBCBT to detect vulnerabilities on MAC layer. It generates elaborated frames test cases transmitting to the target nodes. Based on ZBCBT, we present a ZigBee Border Conditions Algorithm ZBCA and a hostile test framework HTF for better test performance. Comparing with Random Testing RT or fuzz methods, ZBCA tests border values of every field in the frame to improve the efficiency. HTF is a unique framework that ZBCBT simulates an attacker is utilised for further tests. The experimental results, including one frame triggers the network crash, have proved ZBCBT's effectiveness. Thus, by using ZBCA and HTF, this novel vulnerability detection method is a significant add-on approach for ZigBee security.

WEB APPLICATION VULNERABILITY DETECTION BASED ON REINFORCEMENT LEARNING

To solve the problem of low crawling yield and low detection efficiency in web applications security detection, we propose a web application security vulnerability detection method based on Q-learning. We present a strategy of form focused crawling (QLC) which uses Q-learning algorithm to increase the crawling yield and detection efficiency. In the learning algorithm, we present the method of combining immediate rewards and future rewards to evaluate and optimize the learning rules. Simulating web attacking and analyzing the data of response are used to detect security vulnerabilities, and rich attacking vectors ensure the improvement of detection accuracy. Finally, through effective training of the reinforcement learning the rules, a series of experimental results verify the effectiveness of the method we proposed in this paper.

Auditing the XSS defence features implemented in web application programs

Cross site scripting (XSS) vulnerability is mainly caused by the failure of web applications in sanitising user inputs embedded in web pages. Even though state-of-the-art defensive coding methods and vulnerability detection methods are often used by developers and security auditors, XSS flaws still remain in many applications because of (i) the difficulty of adopting these methods, (ii) the inadequate implementation of these methods, and/or (iii) the lack of understanding of XSS problem. To address this issue, this study proposes a code-auditing approach that recovers the defence model implemented in program source code and suggests guidelines for checking the adequacy of recovered model against XSS attacks. On the basis of the possible implementation patterns of defensive coding methods, our approach extracts all such defences implemented for securing each potentially vulnerable HTML output. It then introduces a variant of control flow graph, called tainted-information flow graph, as a model to audit the adequacy of XSS defence artefacts. The authors evaluated the proposed method based on the experiments on seven Java-based web applications. In the auditing experiments, our approach was effective in recovering all the XSS defence features implemented in the test subjects. The extracted artefacts were also shown to be useful for filtering the false-positive cases reported by a vulnerability detection method and helpful in fixing the vulnerable code sections.

A Static Vulnerabilities Detection Method Based on Security State Tracing and Checking

The main problem of existing static vulnerability detection methods based source code analysis is their high false positive and false negative rates.One main reason is lack of accurate and effective identification and analysis of security-related program elements,e.g.data validation checking,tainted data source,etc.A static vulnerability detection method based on data security state tracing and checking is proposed.In this method,the state space of state machine model is extended;the security state of a variable is identified by a vector that may correspond to multiple security-related properties rather than by a single property;Fine-grained state transition is provided to support accurate recognition of program security-related behaviors;The recognition of validation checking is introduced in vulnerability state machine to reduce false positives;and a systematic discrimination mechanism for tainted data is constructed to prevent false negatives result from neglecting tainted data sources.The experimental results of a prototype system show that this method can effectively detect buffer overflow and other type's vulnerabilities in software systems,and with obviously lower false positive than existing mainstream static detection methods and avoid some serious false negatives of these methods.