mini-ELSA: using Machine Learning to improve space efficiency in Edge Lightweight Searchable Attribute-based encryption for Industry 4.0

TL;DR

This paper introduces mini-ELSA, an improved edge encryption method that leverages machine learning to reduce storage and network load in Industry 4.0 applications, enhancing efficiency and processing speed.

Contribution

The paper presents mini-ELSA, an enhancement of the original ELSA method, integrating ML to minimize lookup tables and data records for better edge computing performance.

Findings

Storage reduced by 21%

Execution time improved by 1.27x

Effective in power plant dataset

Abstract

In previous work a novel Edge Lightweight Searchable Attribute-based encryption (ELSA) method was proposed to support Industry 4.0 and specifically Industrial Internet of Things applications. In this paper, we aim to improve ELSA by minimising the lookup table size and summarising the data records by integrating Machine Learning (ML) methods suitable for execution at the edge. This integration will eliminate records of unnecessary data by evaluating added value to further processing. Thus, resulting in the minimization of both the lookup table size, the cloud storage and the network traffic taking full advantage of the edge architecture benefits. We demonstrate our mini-ELSA expanded method on a well-known power plant dataset. Our results demonstrate a reduction of storage requirements by 21% while improving execution time by 1.27x.