Protection Scheme for IoT Device using Introspection

Abstract

A lot of data is being produced with the advances in hardware, software, and network, such as ubiquitous, home network, and M2M. The need is to provide customers with better services using this data. IoT (Internet of Things) is an emerging paradigm that is playing a significant role in this regard. The users get various IoT services produced by collecting and processing data through the interaction of IoT devices. However, the collected data from this process is closely related to the user’s privacy and personal profile. Therefore, the protection of users’ private data is a critical issue in an IoT environment. This thesis proposes a method of protecting the device and the smart sensor by utilizing Introspection in a gateway using virtualization technology in order to detect malicious actions and protect the user`s privacy. For smart sensor and devices from which Introspection cannot directly acquire data, a driver-based light Agent is used to detect malicious activities. After that, the information is sent to the Introspection module in order to detect various malevolent activities. Various OS related functionalities, that are operating inside the gateway for IoT service, can also be analyzed using Introspection. The performance of proposed method was analyzed through an experiment utilizing methodology based on X86 ARM processors. The evaluation results verifies that the load of Introspection was low compared to the Agent and also the power consumption was low as well. Therefore, an improved IoT environment using the suggested method can increase the credibility of the acquired information.