DDoS Attack Detection in IoT Environment Using Crystal Optimized Deep Neural Network

Internet of Things (IoT) is the interconnection of many devices through the internet for different real-time applications. One of the major issues of IoT is Distributed Denial of Service attack(DDoS) and many research works have been carried out to circumvent the DDoS attack; however, they failed to attain the accurate classification of DDoS and normal traffic. In context with this, we propose a novel Deep Neural Network (DNN) based Crystal Search algorithm (CSA) (DNN-CSA). The modified Multilayer Preceptor (MLP) based DNN enhances the classification outcomes. Prior to classification the data are collected by sniffer tool and preprocessed using the min-max approach. Experimental analyses are carried out to analyze the performance of our proposed approach and the results are compared with other state-of-art works. The proposed methodology offers better detection accuracy, precision, recall, and F1-score for DDoS attack detection and also effective results in terms of throughput, energy consumption, and memory utilization.