Abstract
This paper proposes an Optimistic Byzantine Fault Tolerance (oBFT) consensus mechanism specifically designed for Directed Acyclic Graph (DAG)-based blockchains in Industrial Internet of Things (IIoT) networks. While DAG-based architectures offer high scalability and low transactional costs, they often suffer from slow finality and vulnerability to double-spending or parasitic chain attacks under high network latency. Traditional BFT consensus mechanisms mitigate these security risks but impose unsustainable communication overheads (scaling quadratically with the number of nodes) on resource-constrained IIoT devices. To address this trade-off, our proposed oBFT protocol leverages an optimistic execution path that assumes honest node behavior under normal conditions, reducing communication complexity to linear. Upon detecting conflicting transactions or malicious activity, the protocol seamlessly transitions to a robust, multi-stage Byzantine agreement fallback path. We evaluate the performance of oBFT using a simulated IIoT environment with varying network topologies and Byzantine node ratios. The experimental results demonstrate that oBFT achieves up to an 85% reduction in consensus latency and a 60% improvement in throughput compared to traditional Practical Byzantine Fault Tolerance (PBFT) and standard DAG selection algorithms, while maintaining strong security guarantees against up to 33% malicious actors. This mechanism presents a viable pathway for secure, real-time data integrity and transactional security in decentralized industrial environments.
