Evaluating and Optimizing Cryptographic Offloading for IoT Devices: Attribute-Based Encryption Case Study

Sunanda Roy, Thomas Crowley, Brian L. Mark, Kai Zeng, Khaled N. Khasawneh

Abstract: Computational offloading can provide significant latency and energy savings. However, most computational offloading protocols have been theoretically proposed, with some empirical evidence supporting their effectiveness. In this paper, we implement and critically evaluate a reference computational offloading protocol for attribute-based encryption on small-scale IoT devices at the network's edge. Our analysis led to the development of an enhanced version of the protocol that further reduces latency and energy usage without compromising the original protocol's security standards. Our results show significant savings of up to 30 × for latency and 32 × for energy consumption using the original protocol, and up to 40 × for latency and 42 × for energy consumption using our proposed enhancement on the end device.