Malware Analysis Of API Calls Using FGPA Hardware Level Security Model

It is clear that something must be done to help the security community assess, test, and control the level of security of embedded systems. Infected targets consume more power than a cleaned target device because the malware performs additional computing tasks that require additional power to the target device’s processor.

The machine learning module can detect malware by analyzing the aggregate power consumption of FPGA hardware. For example, it can detect 60 Hz network activity periods that can correspond to RAM scraping malware. The server can then perform an analysis of power consumption, memory consumption, and memory usage by the API calls.

The first step is to find the leading Application Programming Interfaces (APIs) that lead to the creation of the malware and its execution on the FPGA hardware.

Collecting more API calls that can provide more information about the malware, and finding complex relationships between API calls can improve performance. In addition to analyzing API calls, it is also possible for malware to generate system calls. This approach…