Network assaults and floods, are rising due to the increasing number of IoT devices, posing security and dependability concerns. These attacks cause a denial of service (DoS) and network interruption for IoT devices. Researchers have established multiple methods to track down assaults on weak IoT gadgets. This study provides a deep learning and swarm metaheuristic technique for detecting DDoS assaults in an Internet of Things (IoT) setting. The group search firefly method, a revolutionary improvement on the classic firefly algorithm, is used as a feature selection tool to zero in on the best candidates. In addition, the hyperparameters of the DarkNet are selected and optimized with the help of a suggested technique called the Boosted Reptile Search technique (BRSA) for effective botnet detection. The operatives of the red fox algorithm (RFO) and the triangular mutation operator (TMO) were used to effect this change. The TMO was utilized to progress the misuse phase of the RSA, whereas the RFO was used to improve the exploration phase. The suggested model is verified using the N-BaIoT dataset. Various cutting-edge methods were employed to evaluate and contrast the projected models efficacy. The outcome proves that the recommended approach is superior to alternatives in identifying multiclass botnet assaults.
A. Alanazi, Adwan; D. Alzughaibi, Arwa; Amoudi, Ghada; I. A. Anja, Manahill; O. I. Abaker, Abdelgalal; and Alkhalaf, Salem
"Detection of DDoS Attacks using Enhanced FS with BRSA- based Deep Learning Model in IoT Environment,"
Information Sciences Letters: Vol. 12
, PP -.
Available at: https://digitalcommons.aaru.edu.jo/isl/vol12/iss12/30