Malware Detection in Android IoT Systems Using Deep Learning

Abstract

The Android Operating System (AOS) has been evolving since its inception and it has become one of the most widely used operating system for the Internet of Things (IoT). Due to the high popularity and reliability of AOS for IoT, it is a target of many cyber-attacks which can cause compromise of privacy, financial loss, data integrity, unauthorized access, denial of services and so on. The Android-based IoT (AIoT) devices are extremely vulnerable to various malwares due to the open nature and high acceptance of Android in the market. Recently, several detection preventive malwares are developed to conceal their malicious activities from analysis tools. Hence, conventional malware detection techniques could not be applied and innovative countermeasures against such anti-detection malwares are indispensable to secure the AIoT. In this paper, we proposed the novel deep learning-based real-time multiclass malware detection techniques for the AIoT using dynamic analysis. The results show that the proposed technique outperforms existing malware detection techniques and achieves detection accuracy up to 99.87%.

Finsformer: A Novel Approach to Detecting Financial Attacks Using Transformer and Cluster-Attention

“…Mamunur et al used adversarially retrained samples to reinforce IDS models, ultimately increasing accuracy to over 99%, but the model is prone to overfitting and needs consideration on how to reduce sensitivity to attack behaviors [9]. Waqar Muhammad et al proposed a deep-learning-based malware detection model for Android-based IoT (AIoT) devices to prevent various malware attacks, achieving an accuracy of 99.87% [10]. Sandouka Soha B. et al combined EfficientNet with generative adversarial networks (GANs) for fingerprint presentation attack detection (PAD) and validated the proposed method on the public LivDet2015 dataset, showing that the proposed method outperforms other CNN models [11].…”