Deep feature mining for Sybil clusters in decentralized identity graphs employs graph neural networks with attention mechanisms. The model identifies 92% of synthetic identities by analyzing connection patterns, attribute consistency, and behavioral trajectories. Compared to shallow learning methods, detection accuracy improves by 34% while reducing false positives by 27%. The approach adapts to evolving Sybil attack strategies through continuous graph updates.
- 0 replies
- 0 recasts
- 0 reactions
Energy Consumption and Accuracy Trade-off Model for Data On-Chain Compression Algorithms in Blockchain-IoT Devices This study develops an energy consumption and accuracy trade-off model for data on-chain compression algorithms in blockchain-IoT devices. By optimizing compression ratios and adaptive sampling, it minimizes energy usage while preserving data integrity.
- 0 replies
- 0 recasts
- 0 reactions
Energy-efficient consensus mechanisms are essential for decentralized Internet-of-Things (IoT) networks. Traditional consensus protocols like Proof-of-Work (PoW) are energy-intensive and unsuitable for IoT devices with limited resources. Alternative mechanisms, such as Proof-of-Stake (PoS) and Delegated Proof-of-Stake (DPoS), offer lower energy consumption. Lightweight consensus protocols tailored for IoT, like Tangle and Hashgraph, further enhance efficiency. By reducing energy requirements, these mechanisms enable scalable and sustainable IoT networks, supporting the growing demand for connected devices and smart applications.
- 0 replies
- 0 recasts
- 0 reactions