Intelligent dynamic heterogeneous redundancy architecture for IoT systems

Abstract

The conventional dynamic heterogeneous redundancy (DHR) architecture suffers from the security threats caused by the stability differences and similar vulnerabilities among the executors. To overcome these challenges, we propose an intelligent DHR architecture, which is more feasible by intelligently combining the random distribution based dynamic scheduling algorithm (RD-DS) and information weight and heterogeneity based arbitrament (IWHA) algorithm. In the proposed architecture, the random distribution function and information weight are employed to achieve the optimal selection of executors in the process of RD-DS, which avoids the case that some executors fail to be selected due to their stability difference in the conventional DHR architecture. Then, through introducing the heterogeneity to restrict the information weights in the procedure of the IWHA, the proposed architecture solves the common mode escape issue caused by the existence of multiple identical error output results of similar vulnerabilities. The experimental results characterize that the proposed architecture outperforms in heterogeneity, scheduling times, security, and stability over the conventional DHR architecture under the same conditions.