The design of an intrusion detection system in MANET using the IGWO-ANN classification algorithm

Presently, attacks on the internet are maximised with the internet's enhancement. Intrusion detection system (IDS) is one of the compassionate layers relevant to information protection. Though researchers have found enormous techniques, there are still issues in detecting new intrusions. So, this framework proposes an effective IDS using IQDFA-based feature selection and the IGWO-ANN classification algorithm. Initially, data conversion occurs, where the input data in the form of characters is replaced by the number. Then, to avoid the similar data's training, redundant data is removed. Then, the normalisation occurs, where the feature values are normalised using an average of min and max attribute values. Next, by utilising the IQDFA, the extra features are extracted after the best feature selection. Data classification is conducted using IGWO-ANN. For determining whether the sensor data was attacked or not, the testing of classified data is done. The proposed model's performance analysis exhibited enhanced performance than the prevailing methodologies.