Encryption for edge computing applications

Abstract

As smart sensors and the Internet of Things (IoT) exponentially expand, there is an increased need for effective processing solutions for sensor node data located in the operational arena where it can be leveraged for immediate decision support. Current developments reveal that edge computing, where processing and storage are performed close to data generation locations, can meet this need (Ahmed and Ahmed 2016). Edge computing imparts greater flexibility than that experienced in cloud computing architectures (Khan et al. 2019). Despite these benefits, the literature highlights open security issues in edge computing, particularly in the realm of encryption. A prominent limitation of edge devices is the hardware’s ability to support the computational complexity of traditional encryption methodologies (Alwarafy et al. 2020). Furthermore, encryption on the edge poses challenges in key management, the process by which cryptographic keys are transferred and stored among devices (Zeyu et al. 2020). Though edge computing provides reduced latency in data processing, encryption mechanism utilization reintroduces delay and can hinder achieving real-time results (Yu et al. 2018). The IoT is composed of a wide range of devices with a diverse set of computational capabilities, rendering a homogeneous solution for encryption impractical (Dar et al. 2019). Edge devices are often deployed in operational locations that are vulnerable to physical tampering and attacks. Sensitive data may be compromised if not sufficiently encrypted or if keys are not managed properly. Furthermore, the distributed nature and quantity of edge devices create a vast attack surface that can be compromised in other ways (Xiao et al. 2019). Understanding established mechanisms and exploring emerging methodologies for encryption reveals potential solutions for developing a robust solution for edge computing applications. The purpose of this document is to detail the current research for encryption methods in the edge computing space and highlight the major challenges associated with executing successful encryption on the edge.