AASH: A Lightweight and Efficient Static IoT Malware Detection Technique at Source Code Level

IoT malware applications significantly threaten user privacy and security. Traditionally, IoT developers have focused primarily on hardware, but connectivity requires additional embedded software, usually developed by third-party developers. Unfortunately, third-party code is not always secure and trustworthy, and it frequently contains bugs and malicious code, which leaves IoT devices vulnerable. We propose the AASH technique (IoT Malware Detection) a novel technique that can detect malware at the source code level using the Adler-32 hash function and Fibonacci search. Previously, DROIDMD technique and SQVDT technique have been proposed to detect malware on Android and Linux devices. According to the authors, their schemes are scalable and can be deployed on IoT devices. However, their technique suffers from lower accuracy and takes longer to detect malicious code. The performance measurement shows that our proposed AASH technique is comparatively better than DROIDMD and SQVDT techniques in terms of accuracy and malware detection. AASH is reliable, efficient, and can be deployed on a large-scale level.