IoT Botnet Detection Using Salp Swarm and Ant Lion Hybrid Optimization Model

In the last decade, devices and appliances utilizing Internet of Things (IoT) have expanded tremendously, which has led to revolutionary developments in network industry. Smart homes cities, wearable devices, traffic monitoring, health systems, energy savings are typical IoT applications. The diversity standards, protocols, computational resources makes them vulnerable security attackers. Botnets challenging threats that cause severe Distributed Denial Service (DDoS) attacks. Intrusion detection systems (IDS) necessary for safeguarding Internet-connected frameworks enhancing insufficient traditional countermeasures, including authentication encryption techniques. This paper proposes a wrapper feature selection model (SSA–ALO) by hybridizing salp swarm algorithm (SSA) ant lion optimization (ALO). new can be integrated with IDS components handle high-dimensional space problem detect attacks superior efficiency. experiments were performed using N-BaIoT benchmark dataset, was downloaded from UCI repository. dataset consists nine datasets represent real traffic. experimental results reveal outperformance SSA–ALO compared existing related approaches following evaluation measures: TPR (true positive rate), FPR (false G-mean, processing time, convergence curves. Therefore, proposed serve applications detecting intrusions high true rates reach 99.9% minimal delay even imbalanced intrusion families.