Simple Lightweight Cryptographic Algorithm to Secure Imbedded IoT Devices

Abstract

Improving the voting system has become a widely discussed issue. Paper-based elections are not safe because of the possibility of changing and adding ballots. Consequently, many countries use e-voting systems to ensure security, authenticity and time efficiency. Blockchain e-voting systems can be adopted to reduce fraud and increase voting access from home, especially in pandemics. This paper suggests a blockchain e-voting system that tackles two security and authentication issues. The security has been ensured using hybrid publickey cryptography; the voter information is encrypted using the regional election office elliptic public key, while the homomorphic public supreme election authority encrypts the vote. Using homomorphic encryption for voice enables the calculations of results as the authority encrypts it without revealing the vote itself. Authentication has been improved for home voting by a robust login system. This login system consists of two steps. In the first step, the voter enters the site using his unique QR code number scanned by webcam; in the second step, the system checks the voter's face using a face recognition system by web camera to be routed to the voting page. Voting public keys are also authenticated using a digital certificate schema. The system has been tested to show its efficiency and suitability in block establishment time and the encryption and key generator randomness using NIST tests. The internet of things (IoT) revolution has been sparked by the exponential increase in connected devices caused by recent advances in wireless technology. These embedded devices gather, analyze, and send vast data via the network. Data transmission security is a primary problem in IoT networks. Several lightweight encryption/decryption algorithms have been developed for such resource-constraint devices. One of most effective and fast lightweight encryption algorithms for IoT applications is the Tiny Encryption Algorithm (TEA). TEA has few lines source of codes to implement and based on Feistel structure to provide cryptographic primitive confusion and diffusion features in order to hide statistical aspects of plaintext. However, it is vulnerable to assaults using equivalent and related key attacks. This study suggested modifying TEA by employing a new proposed generating keys function using two Linear Feedback Shift Registers (LFSRs) as a combination to address the security flaw caused by utilizing different keys for each round function. The key sensitivity, Avalanche effect, and a completeness test were used to evaluate its security performance. The key sensitivity of the proposed modified TEA outperforms original TEA by 50.18 % to 44.88 %. The modified TEA avalanche effect outperforms TEA by 52.57 % to 47.69 %, and its completeness test outperforms TEA by 51.75 % to 48.36 %. Experimental results indicates that, the encryption performance of proposed modified TEA is better than original TEA.