Expedient Intrusion Detection System in MANET Using Robust Dragonfly-Optimized Enhanced Naive Bayes (RDO-ENB)

Mobile Ad hoc networks (MANETs) represent dynamic, self-configuring network environments that provide flexible connectivity but are highly susceptible to security threats. Intrusion detection systems in MANETs need to continuously monitor network traffic for potential intrusions and anomalies. This constant monitoring can be energyintensive, requiring network nodes to process, analyze, and transmit data. Excessive energy consumption by IDS can deplete
node batteries quickly, leading to network disruptions. This
research focuses on developing and evaluating an efficient IDS proposed for MANETs called Robust Dragonfly-Optimized
Naive Bayes (RDO-ENB). RDO-ENB operates by fusing the
simplicity and efficiency of the Enhanced Naive Bayes algorithm with the adaptive capabilities of robust Dragonfly Optimization. This synergy enables RDO-ENB to continuously and dynamically adjust its internal parameters, optimizing its intrusion detection performance in real time. It enhances accuracy and reduces false positives, making it proficient in identifying and mitigating intrusions within the complex and ever-evolving environment of MANETs. The dataset employed for evaluation is NSL-KDD, a widely used dataset for intrusion detection. The results of the IDS implementation demonstrate its proficiency in accurately identifying and mitigating intrusions while minimizing false positives and conserving valuable energy resources.