Deep Learning Approach for IoT Malware Detection and Classification based on Squirrel Search Algorithm and Convolutional Neural Network (IMD_SSACNN)

The number of Internet of Things (IoT) devices that are exposed to the public has been rising as more of these devices are connecting to the internet using their default settings. Due to the variety of designs and the IoT's limited computation and storage capacities, it is challenging to implement sufficient security measures, which makes it more susceptible to infection. Accurate IoT malware identification and family attribution are crucial in order to begin implementing attack mitigation/prevention tactics, which is why they are so important in order to reduce the threat. To prevent the risks caused by malicious code, various research has been done on the identification of IoT malware. It might be challenging to recognize the novel variant IoT virus that is being created rapidly, even though existing models might successfully identify hazardous IoT code found through static analysis. This research introduced a novel IoT Malware Detection with a Squirrel Search Algorithm and Convolutional Neural Network (IMD-SSACNN). IoT malware datasets are used to conduct an exhaustive analysis of the suggested technique. IMD-SSACNN is able to lessen the damage that malware infestation brings to IoT devices by examining and analyzing the massive volume of behavior data generated by dynamic analysis. The experimental findings show that the suggested IMD-SSACNN is the preferred approach since it has a greater detection rate than the earlier malware detection algorithms.