A Permissioned Consensus Algorithm Based DAGs-to-Blockchain in Hierarchical Architecture for Decentralized Internet of Things

Abstract

In today’s modern world, distributed ledgers have become dominant in several sophisticated applications to remove the centralized controlling entity (e.g., Cloud Computing). The currents are problematic to meet the dramatic growth of the Internet of Things (IoT), for instance, Blockchain and Tangle. The former is very secure but has a very slow confirmation time, and low throughput. The latter has more throughput; however, sometimes records are getting left behind. On the other hand, the existing consensus algorithms that determine who can add new records into a ledger, are particularly energy-intensive, unfair, or deemed infeasible for resource-constrained Things. In this paper, a novel consensus algorithm based on Directed Acyclic Graphs (DAGs)-to-Blockchain is proposed in horizontal architecture to mitigate the scalability issues, in which every limited number of nodes are grouped. It consists of five permissioned nodes of different responsibilities: (i) Things send data from sensors with signature to Edge. Every interval, (ii) each Edge node chooses a random number, and auto-decrement it. The first node reaches zero, becomes the winner that can add the verified data temporally to the DAG ledger (Cobweb) depends on uniform distribution. The rest of the nodes who are waiting for the leader, rate it based on its work (rate scores from 1-to-5) after accepting Cobweb from it. (iii) Fog node is chosen randomly from Edge to send Cobweb periodically with authentication to reduce the required storage. (iv) Cloud nodes receive and add it permanently to the Blockchain ledger with global accessibility. (v) Queen node is located between Fog and Cloud, which monitors the network. The proposal is emulated and evaluated in the real-time scenario.