Fuente:
PubMed "smart farming"
Sci Rep. 2026 May 11. doi: 10.1038/s41598-026-52193-4. Online ahead of print.ABSTRACTThe integration of Internet of Things (IoT), blockchain, and artificial intelligence (AI) holds great promise for precision agriculture, yet challenges remain in secure data acquisition, authenticated device interactions, and transparent decision-making. This paper presents RiceBlock, a blockchain-IoT framework that secures environmental data collection and automates agronomic decisions for precision rice farming. SHA-256 hashing and AES-256-GCM encryption were employed in the framework to ensure data integrity and confidentiality. In addition, mutual authentication between sensor nodes, cluster heads (CHs), and user mobile devices (UMDs) is achieved via Ethereum digital signatures and timestamp verification. A set of Solidity smart contracts including [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text] were used in the system to govern device identity, data provenance, and rule‑based AI decisions under Proof‑of‑Authority (PoA) consensus. The system was evaluated on a testbed of 20 sensors across five zones, generating 3000 transaction records. RiceBlock achieves a mean transaction latency of 5.73 s, throughput of 22 tx/s, and an authentication success rate of 98.2%, with low false-acceptance (1.8%) and equal-error (2.1%) rates. Decision accuracy is confirmed by an F1-score of 0.93 and a ROC-AUC of 0.96, demonstrating resilience to replay and forgery attacks. The framework provides a scalable, privacy-preserving architecture that bridges blockchain-based trust with actionable AI-driven agricultural intelligence.PMID:42115703 | DOI:10.1038/s41598-026-52193-4