An Optimized Static Propositional Function Model to Detect Software Vulnerability

Abstract

Due to the lack of appropriate theory to accurately characterize vulnerabilities, the current static detection technologies have two key challenges, i.e., limited applicability, and the problem of state space explosion. In this paper, we put forward a static detection model based on the proposition function. Furthermore, a new program intermediate representation called Vulnerability Executable Path Set (VEPS) is proposed to optimize our model which compresses the program state space distinctly. In addition, in order to confirm the reliability of the static detection model, we conduct three terms of contrast experiments to estimate the results with the vulnerability disclosed by NIST. The results obtained from extensive experiments show that the proposed model effectively detects more Wireshark bugs than NIST, and reveals a higher detection efficiency than FindBugs.

A systematic literature review of software vulnerability detection

“…Wibowo et al (2017) investigated the Architectural Vulnerability Factor (AVF) of all major in-core memory structures of an out-of-order superscalar processor while Sultana and Williams (2017) used micro patterns detect vulnerability in software. Ziems and Wu (2021) and Paradis et al, (2018) modelled test as a source code and used it for software vulnerability detection in natural language processing (NLP), Han et al (2019) proposed a static detection model, while Choi et al (2020) developed Cyber-Physical Inconsistency to target vulnerability detection in Robotic Vehicles (RVs).…”

An Environment-Specific Prioritization Model for Information-Security Vulnerabilities Based on Risk Factor Analysis

“…The number of vulnerabilities is greater because Shodan knows more specific organization data and can relate a more significant number of known vulnerabilities. Because Shodan is a secure software and its documentation only describes the variable "tag" for the query processes and not in the response processes, we have limited ourselves to verify its existence in order to assign a risk value for the prioritization process [21,40]. Since the quality of information exposed in Shodan provides more significant value in knowledge and investigation for cybercriminals [20,22], the risk increases when this variable called "tags" has some content.…”

Vulnerability and Risk Analysis Methods and Application in Large Scale Development of Secure Systems

“…A review of existing related work in this area indicates that vulnerability detection is still a challenging area for software developers. Han et al [76] proposes that one of the reasons that the currently available static 7. Related work -SRA detection technologies have limited applicability and the problem of state space explosion (the exponential increase in the size of system state space with growing number of state variables) is the lack of appropriate theory to accurately characterize vulnerabilities.…”

“…We found the basic idea of describing a vulnerability in terms of its preconditions, characteristics, and decision rules to be similar to VCGs, where every path from the top cause in the graph, down to the vulnerability is describing a possible execution path that leads to the existence of a vulnerability. Han et al [76] refers to our vulnerability modeling technique as a structured method for analyzing and recording the causes of software vulnerabilities. VCGs are described as one of the detection frameworks for vulnerability positioning and discrimination.…”