UCFL: User Categorization using Fuzzy Logic towards PUF based Two-Phase Authentication of Fog assisted IoT devices

Abstract

Internet of Things (IoT) makes the daily life easier by increasing the connectivity between the devices. The increase in number of connected devices makes the network vulnerable to various threats. The resource constraint sensor nodes added more challenge for the researchers to implement security schemes in IoT network. In this paper a three layer framework for IoT environment is designed where the users of first layer are intended to collect data from the sensor nodes of third layer through the fog devices that are belongs to second layer. The resource constraint sensor nodes are released from checking the authenticity of users to save their energy. Thus some high capacity nodes, called fog devices, are involved to check the authentication of users. The users are categorized as high, medium or low trusted by using fuzzy logic. Further it delves for the authentication of low and medium trusted user by implementing a lightweight two phase authentication scheme. Physical Unclonable Function (PUF) is used to generate unique identity for individual users and to create Challenge Response Pairs (CRPs). The performance is evaluated by using Raspberry Pi as a user and a laptop as a fog device. The simulation results show that the proposed scheme outperforms the existing schemes in terms of average response time, handshake duration, memory utilization and communication cost. Moreover this scheme is resilience against replay attack, DoS attack, false data injection and man in the middle attack.