LAA-D: Lightweight Authentication and Access Control Mechanism with Dual-Data Storage in Cloud-Internet of Things System Using Blockchain

Abstract

Internet of Things (IoT) applications are popularly involved in day-to-day life. The increase in utilization leads to an increase in network traffic. The incoming users have different intentions in the network and hence security is essential. The data user accesses the data in the cloud that is collected from IoT devices. A large-scale IoT environment has challenges in the provisioning of security as well as the management of access control mechanisms. The problem is a generation of policies and authenticating devices with minimum credentials. In this paper, Blockchain-based decentralized authentication and access control systems are designed. The process of authentication is conducted for the data owner and data user by considering identity, device type, IP address and signature, PUF, and biometric respectively. PUF stands for Physical Unclonable Function, which is a hardware-based security feature that generates a unique identifier for a device based on its physical properties, SALSA20 and PRESENT are encryption algorithms used in the proposed system to encrypt data chunks. SALSA20 is a stream cipher that generates a keystream to encrypt data, while PRESENT is a block cipher that encrypts data in fixed-size blocks These authentication credentials are managed in the blockchain. The credentials are stored in encrypted form using the Key schedule PRESENT algorithm. In the authentication of data users, the number of credentials is selected using fuzzy logic that improves security. To assure data storage security, the data is split into two chunks, and it is encrypted using SALSA20 and PRESENT algorithm. The proposed model is developed in an ifogsim simulator, and the performance metrics are evaluated in terms of authentication time, storage efficiency, running time, throughput, latency, and blocksize.