Modeling and analyzing cascading failures for Internet of Things

Network survivability is one of the key challenges that the Internet of Things (IoTs) has to deal with, and cascading failures are one of the main bottlenecks affecting the network survivability of the IoTs. In this work, we fully considered the real characteristics of IoTs systems (i.e., data aggregation and link heterogeneity), and proposed a realistic cascading model based on the layered architecture of the IoTs. In this model, the cascading process of the IoTs is driven by the overload events of relay nodes, base stations and communication links. To help the IoTs improve network survivability, a load-oriented layout scheme for base stations is presented. Through extensive simulations, the soundness of the proposed cascading model and the effectiveness of the proposed layout scheme have been verified, and some meaningful findings are obtained: there is a fully tolerance parameter space for the capacity expansion of network components, and when the expansion coefficients are in this space, the removal of a single network component will not trigger cascading failures; compared with removing other types of network components individually, removing base stations is more likely to trigger cascading failures; in addition to capacity expansion, deploying more base stations can also significantly improve network survivability. (C) 2020 Elsevier Inc. All rights reserved.

Automated summary

The study proposed a realistic cascading model based on the layered architecture of IoTs, and presented a load-oriented layout scheme for base stations to improve network survivability. Experiment results verified the validity of the model, showing a tolerance parameter space for capacity expansion of network components, the higher likelihood of triggering cascading failures by removing base stations, and the significant improvement in network survivability by deploying more base stations.