Obfuscation Techniques against Signature-Based Detection: A Case Study
Gerardo Canfora, Andrea Di Sorbo, Francesco Mercaldo, Corrado Aaron Visaggio
Abstract
Gerardo Canfora, Andrea Di Sorbo, Francesco Mercaldo, Corrado Aaron Visaggio
Abstract
Android malware is increasingly growing interms of complexity. In order to evade signature-based detection, which represents the most adopted technique by current antimalware vendors, malware writers begin to deploy malware with the ability to change their code as they propagate.In this paper, our aim is to evaluate the robustness of Android antimalware tools when various evasion techniques are used to obfuscate malicious payloads. To support this assessment we realized a tool which applies a number of common transformations on the code of malware applications, and applied these transformations to about 5000 malware apps. Our results demonstrate that, after the code transformations, the malware is not detected by a large set of antimalware tools,even when, before applying the transformations, malware was correctly identified by most antimalware tools. Such outcomes suggest that malware detection methods must be quickly re-designed for protecting successfully smart devices.
OpenAlex reports 57 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Android malware is increasingly growing interms of complexity. In order to evade signature-based detection, which represents the most adopted technique by current antimalware vendors, malware writers begin to deploy malware with the ability to change their code as they propagate.In this paper, our aim is to evaluate the robustness of Android antimalware tools when various evasion techniques are used to obfuscate malicious payloads. To support this assessment we realized a tool which applies a number of common transformations on the code of malware applications, and applied these transformations to about 5000 malware apps. Our results demonstrate that, after the code transformations, the malware is not detected by a large set of antimalware tools,even when, before applying the transformations, malware was correctly identified by most antimalware tools. Such outcomes suggest that malware detection methods must be quickly re-designed for protecting successfully smart devices.
Key concepts: Malware, Cryptovirology, Computer science, Obfuscation, Android malware, Android (operating system), Evasion (ethics), Mobile malware