The Internet has extended by including a large number of IoT devices implemented recently. These devices are smarterdue to the stronger computational capability and the interconnection through Internet therefore can deal with much morecomplicated tasks. On the otherside, there are also more chances for attackers to threaten these things. In this paper, we propose anovel light-weight approach for detecting DDos malware in IoT environments, through malware image and a light-weightconvolutional neural network image classifier. The results show that the proposed system can achieve 94.0% accuracy for theclassification of goodware and DDoS malware..
展开▼