Intrusion Detection in IoT Platform Using Tuna Swarm Optimization with Long Short-Term Memory
Sanjeevini S Harwalkar, A. H. A. Hussein, Bura Vijay Kumar, Mohammed Ihsan Habelalmateen, R. Mary Victoria · 2023
The Internet of Things (IoT) is a dynamic and delightful research field in this emerging technology. It can be globally connected with many IoT devices and exchange a large amount of data. However, the threats also developed and misguided the entire network behaviour. This article proposes an Intrusion Detection System (IDS) using the proposed Long Short-Term Memory (LSTM) along with the Tuna Swarm Optimization (TSO) to fine tune the hyperparameters and helps to enhance the detection accuracy of attacks that takes place in IoT environment. The data is obtained from NSL-KDD data set and it is pre-processed using Synthetic Minority Over-Sampling Technique (SMOTE) and it is fed into the stage of feature extraction which is performed using the Recursive Feature Elimination (RFE). After this, the feature selection is performed using optimization-based feature selection approach based on Moth Flame Optimization (MFO) and finally, the detection of attack takes place with the help of TSO-LSTM where the results exhibited by the LSTM are fine-tuned with the help of TSO which helps to enhance the detection accuracy of the model. The detection accuracy of the proposed TSO-LSTM is 99.98% which is comparably higher than the existing approaches such as DBN, SMO-HPSO, CNN-LSTM and BFO-RF with detection accuracy of 99.79%, 99.17%,98.8% and 99.96% respectively.