Multi-Parameter and Time Series Based Trust for IoT Smart Sensors

Zhi-Ge He

《International Journal of Network Security》 22卷4期 (2020/07) Pp. 589-596

https://doi.org/10.6633/IJNS.202007_22(4).06

摘要

The Internet of Things, or IoT has achieved much attention in the past few years with many concrete applications. Among various IoT components, smart sensors play a vital role for things' tracking and monitoring, but due to the absence of centralized administration, those sensors may encounter various security issues which hinder IoT further development. Trust computing provides dynamic behavior perceiving capability and can take precautionary measures against malicious actions. In this study, unlike traditional binary parameter trust, we first propose a multi-parameter trust computing method so that trust states can be more accurately and practically described, then according to the theory of time series, a favorable trust data sequence and an unknown trust data sequence are generated so that nodes' malicious actions can be observed and detected from the context of a time period. Simulation results show that the proposed method can generate a fast detection of malicious nodes, a higher data packet delivery ratio, and a more trusted network environment ideal for transactions among sensor nodes.

關鍵字

IoT ； Multi-Parameter Trust ； Smart Sensors ； Time Series