Towards Machine Learning Based Intrusion Detection in IoT Networks

The Internet of Things (IoT) integrates billions self-organized and heterogeneous smart nodes that communicate with each other without human intervention. In recent years, IoT based systems have been used in improving the experience many applications including healthcare, agriculture, supply chain, education, transportation traffic monitoring, utility services etc. However, node heterogeneity raised security concern which is one most complicated issues on IoT. Implementing measures, encryption, access control, authentication for devices are ineffective achieving security. this paper, we identified various types threats shallow (such as decision tree (DT), random forest (RF), support vector machine (SVM)) well deep learning (deep neural network (DNN), belief (DBN), long short-term memory (LSTM), stacked LSTM, bidirectional LSTM (Bi-LSTM)) intrusion detection (IDS) environment discussed. performance these models has evaluated using five benchmark datasets such NSL-KDD, IoTDevNet, DS2OS, IoTID20, Botnet dataset. metrics Accuracy, Precision, Recall, F1-score were to evaluate shallow/deep IDS. It found IDS outperforms detecting attacks.