Machine-learning approach for detecting cyberattacks in Medical Internet of Things

To secure wearable sensor data sharing through the cloud (Think speak server) by utilizing the merits of Medical Internet of Things (IoMT), an encryption technique as well as machine learning-based attack detection is developed. The usage of wearable Internet of Things (IoT) devices is escalating day by day and the risks in cyber threats also raising exponentially. While considering the privacy and security of the data over internet, all possible threats and attacks have to be considered. The main attacks presented in wireless networks are data injection attack, Denial of Service (DoS) attack and Zero day attack. The data is protected using the advanced encryption standard (AES) encryption technique, while mentioned attacks are detected using the capabilities of artificial neural networks (ANNs) in the technique explained. The proposed method is capable of detecting different attacks and ensures more security to the data over internet. The method can be used in many other applications such as vehicular IoT, wireless sensor networks (WSNs) and Industrial IoT.