Secured Lightweight Encryption using Time-based Singular Value Decomposition (SVD) for Cyber-Physical Systems

Abstract Cyber-physical Systems (CPS) represent a pioneering category of IoT systems seamlessly integrating computing, communication, and physical components. These systems interact with the physical world, gather sensor data, interpret information, and regulate processes, facilitating real-time responses and efficient resource utilization across diverse domains such as healthcare, Industry 4.0, autonomous vehicles, and smart grids. However, the implementation of CPS presents multifaceted challenges including security and privacy, in-teroperability, scalability, real-time operation, data management, and energy efficiency. Security, in particular, emerges as a paramount concern due to the vulnerability of CPS to cyberattacks, which can yield severe real-world consequences including physical harm, environmental damage, or financial losses. Protecting sensitive data transmitted within CPS is imperative to thwart unauthorized access, data breaches, and malicious manipulation, especially considering their role in controlling critical infrastructure. Compliance with cybersecuri-ty regulations is essential, necessitating robust security measures throughout the CPS supply chain. Resilience against equipment failure, natural disasters, and cyber threats can be enhanced through redundancy, failover mechanisms, and incident response strategies. Given the computational constraints of CPS devices , the adoption of lightweight encryption algorithms is advocated to ensure data security without overwhelming limited processing power, memory, and energy resources. Effectively managing and upholding CPS necessitates a proactive approach towards implementing stringent security measures and remaining vigilant against evolving threats. In this research, a novel CPS encryption using a Time-based SVD algorithm is developed and tested on various current-generation hardware devices like Node MCU, Arduino, Raspberry Pi, and STM32WLE series controllers used in the Internet of Things (IoT) domain. The proposed security algorithm is compared on various attributes like latency, execution time, and energy consumption for executing the algorithm, and observed from the results that Node MCU outperforms the rest.