Seamless Human Impedance-based IoT Authentication with Machine Learning Techniques

A robust authentication system is seen as a solution to ensure security within IoT devices since such devices are vulnerable to attack and data insecurity. In addition, the system could potentially bring some unpleasant experience to the user. Especially, if it is implemented in an everyday-used device. As the ability of the IoT devices to capture the medical data and human behavior series is improved by the time, this advantage creates an opportunity to enhance and diversify the biometric-based authentication. The growth of artificial intelligence and its integration to the authentication-related research create a positive trend. Human impedance is one of the biodata generated from the human body and it has been used for certain medical analysis. This work provides an IoT authentication system based on machine learning. The system utilizes human impedance as a user identifier and machine learning to create an identification machine to determine the pattern similarity between the input and the existing model with a mechanism to preserve its performance. This paper also gives an insight into how human impedance can be incorporated as a biometric identifier for the authentication system.