IoT Based Intrusion Detection System for Healthcare Using RNNBiLSTM Deep Learning Strategy with Custom Features

Abstract Internet of Things (IoT) devices exchange information directly between devices. They are more prone to vulnerability because of the design of the network layer in its architecture and also connected to the internet 24X7. IoT-based smart healthcare devices like patient monitoring cameras in hospital create life-saving data that must be shielded from intruders. Effective intrusion detection is required to safeguard sensitive private data before assault takes place due to the humongous data created by the IoT. This work proposes a 5-layered framework to find intrusion in large datasets. This work uses constructing new custom features to increase the learning rate and to reduce imperceptions during learning by the machine model. The proposed ACAAS algorithm obtains significant features and Recurrent Neural Networks with Long Short-Term Memory in both directions (RNNBiLSTM) is used to identify the assault to optimize the prediction performance accuracy by using the IoTID20 dataset to protect IoT networks. The experiment results provided Accuracy Rate of 99.16%, Error Rate of 0.008371%, Sensitivity Ratio of 99.89% and Specificity Ratio of 98.203% for IoTID20 with custom features using RNNBiLSTM. The obtained high accuracy rate shows the effectiveness of the system in protecting the network from intruders.