Performance Measures of Behavior-Based Signatures: An Anti-malware Solution for Platforms with Limited Computing Resource

摘要

The signature-based malware-detection method is the most popular one used in anti-malware software. However, given advanced malware capabilities, the database of traditional signature-based antimalware software is becoming bloated to support identification of every variant. The increase in signatures slows the detection process and, in some cases, exceeds the resource availability of the platforms that need it most. With the expansion of the smaller platforms with limited computing resources, such as some mobile devices and various types of sensor networks, including Internet-of-Things (IoT), anti-malware's capability needs to be refined to support these platforms. Behavior-based signatures might provide that much-needed reduction in the number of signatures found in a signature set while retaining the full spectrum of malware variants.