Nowadays, Distributed Denial of Service (DDoS) attacks have proved to be uncontrolled and arrive in different patterns and shapes. Therefore, it is hard to identify and resolve the preceding solutions. A DDoS attack is a mischievous try to interrupt the usual traffic of a target server, network, or service by overcoming the aim or its nearby framework with an overflow of Internet traffic. Then, numerous classification methods are applied in many research and are targeted to identify and resolve the attack of DDoS. DDoS attacks are implemented simply by utilizing the network's flaws and by making desires for software services. The real-time detection of attacks is challenging to identify and alleviate. However, this solution is valuable as these attacks may lead to significant problems. Several deep learning (DL) methods are advanced to locate and analyze DDoS attacks. This study proposes a novel Intelligent Framework for Attack Detection Using Salp Swarm-Based Feature Selection and Deep Learning Architecture (IFAD-SSFSDLA) model. This paper aims to deliver a real-time DDoS attack detection system utilizing advanced optimization algorithms. As a primary step, the IFAD-SSFSDLA technique utilizes min-max normalization for the data pre-processing to transform, clean, and organize raw data into the structured pattern. In addition, the Salp swarm algorithm (SSA) is employed in the feature selection process to detect and maintain the most significant features to improve the model performance. The IFAD-SSFSDLA model implements the temporal convolutional network (TCN) method for the attack classification. To exhibit the enhanced performance of the IFAD-SSFSDLA model, a comprehensive experimental analysis is conducted using CIC-IDS-2017 and Edge-IIoT datasets. The performance validation of the IFAD-SSFSDLA model portrayed superior accuracy values of 99.56% and 99.65% over existing techniques under dual datasets.
Alamro et al. (Tue,) studied this question.