Abstract:
Fog computing has transformed the Internet of Things by delivering computing facilities of the
Cloud at the edge of the network. Compared to Cloud, Fog offers various advantages, including
reduced latency and enhanced reliability. The Fog architecture is very dynamic in nature and
its topology continuously changes as new objects enter and leave the network. This opens the
door for malicious objects to infiltrate and disrupt the network. As the Fog-based IoT network
expands rapidly, effective security in such a scenario becomes critical. Trust-based approaches
have recently piqued the academic community’s interest to mitigate the security concerns of the
Fog-based IoT. Trust among objects is vital because it allows them to distinguish and govern information
exchange in the network. However, in the Fog-based IoT network context, which is a
fundamental aspect of trust, has not been given sufficient attention yet. Thus, we propose in this
paper a context-oriented, multi-source, similarity measures-based trust management system for
Fog-based IoT. Our trust management model effectively incorporates context in the trust calculation
process with respect to both servers and recommenders. By utilizing Bayesian inference
and similarity scores to filter contextually similar service providers and recommenders respectively,
our proposed trust management system is effective in calculating a context aware trust
score.
