Malware Detection in Internet of Things Devices Through the Utilization of Deep Sparse Autoencoders and Signal Processing Methods

As a result of the widespread use of the Android Mobile operating system, there has been a rise in the number of malicious software applications that are designed to do damage to those who use this system. Consequently, a great number of investigations have been carried out to identify harmful software for Android. The categorization of software that is known to be hazardous according to the families to which it belongs is also highly essential within the context of the security of the Android operating system. This classification is in addition to the classification of Android software as either harmful or harmless. ESD, a signal processing function, was integrated with deep sparse autoencoders in this work to identify malicious software that is present in Internet of Things devices. The suggested technique demonstrated an accuracy rate of 99.0 percent when compared to the number of research studies conducted in the area of malware identification.