Intrusion detection system using RBPSO and fuzzy neuro-genetic classification algorithms in wireless sensor networks

In the recent years, intrusion detection systems (IDSs) are used to detect malicious attacks in WSNs. Most of the IDSs are developed using two phases namely, a feature selection phase for selecting only the important features and classification phase to classify the intrusions. The existing works on IDS are focusing on providing security to wired networks, databases and operating systems. However, WSNs need a different type of IDS due to the mobility of nodes, limited energy in nodes and necessity for base station to collect the data. Therefore, we propose a new IDS that is more suitable for securing the communication in WSN. In this paper, a new feature selection algorithm based on PSO is proposed called rule-based PSO (RBPSO) and fuzzy neuro-genetic classification algorithm (FNGCA) for securing WSN. The major uses of this proposed model include increased detection rate and reliable communication with reduction in energy consumption and delay.