Smart Contracts for Certified and Sustainable Safety-Critical Continuous Monitoring Applications

Abstract

Monitoring applications are increasingly important to enable predictive maintenance and real-time anomaly detection in industrial and civil safety-critical infrastructures. Typical monitoring pipelines consist of a sensor network that collects and streams IoT data toward a cloud infrastructure that provides storage, visualisation and data analytic capabilities. However, since critical data generated must be often retained for regulatory and tracking purposes, cloud storage requirements become poorly sustainable when dealing with critical infrastructures that have to remain operative for decades while supporting lifelong continuous monitoring. While policies can be applied to remove redundant or outdated information, anti-tamper mechanisms are required to guarantee that data modifications are not driven by malicious intents to alter recorded data. This work presents a blockchain-based framework for continuous monitoring applications enabling certified removal of IoT data in safety-critical databases. The framework allows for the deployment of data-evaluation policies to identify redundant/outdated measurements flowing in the database and, therefore, mark them as eligible for removal. The novelty of our approach stands in the implementation of the data-evaluation policy as a smart contract. Furthermore, the use of a blockchain ensures that critical database operations (like removal) are tamper-proof and compliant with the guideline determined by system stakeholders. We demonstrate the effectiveness of the proposed framework in a real case study using accelerometer data of a bridge monitoring application, and we characterise the overhead of transactions to the blockchain.

N. Elia—Acknowledges support from TIM S.p.A. through the PhD scholarship.