Social learning for resilient data fusion against data falsification attacks

TL;DR

This paper introduces a social learning-based distributed data fusion scheme for IoT networks that enhances resilience against data falsification attacks by mimicking social network behaviors.

Contribution

It proposes a novel social learning approach for distributed decision-making in IoT, improving security against compromised nodes compared to traditional centralized methods.

Findings

Distributed social learning improves attack resilience

Local actions influence global network behavior

Scheme maintains performance even with many compromised nodes

Abstract

Internet of Things (IoT) suffers from vulnerable sensor nodes, which are likely to endure data falsification attacks following physical or cyber capture. Moreover, centralized decision-making and data fusion schemes commonly used by these networks turn these decision points into single points of failure, which are likely to be exploited by smart attackers. In order to face this serious security thread, we propose a novel scheme that enables distributed data aggregation and decision-making by following social learning principles. Our proposed scheme makes sensor nodes to act resembling the manners of agents within a social network. We analytically examine how local actions of individual agents can propagate through the whole network, affecting the collective behaviour. Finally, we show how social learning can enable network resilience against data falsification attacks, even when a significant portion of the nodes have been compromised by the adversary.