Title

Private Matching and Set Intersection Computation in Multi‐Agent and Industrial Control Systems

Document Type

Conference Proceeding

Publication Date

4-4-2017

Publication Title

Proceedings of the Cyber and Information Security Research Conference

DOI

10.1145/3064814.3064817

ISBN

978-1-4503-4855-3

Abstract

Distributed autonomous systems that rely on dataset matching and set intersection computation for decision making capabilities are vulnerable to datasets poisoning attacks. Among these systems, Industrial Control Systems (ICS) operating on critical infrastructures. Attacker with a compromised Programmable Logic Controllers (PLCs) can take advantage of the PLC-to-PLC information sharing process to construct and inject anomalous data that target the result of dataset matching and set intersection computation and hence bring the process operations into unstable state. We introduce a protocol that utilizes secure hamming distance computation from oblivious transfer to compute a joint set between two system's agents that hold private input datasets of length n. The proposed protocol achieves full security in the semihonest model.

Share

COinS