Scholarlink

<p>The convergence of Artificial Intelligence (AI) and Blockchain technology has significantly advanced smart agriculture by enabling secure, transparent, and efficient data transactions. Even with these technological improvements, current blockchain security methods like Proof of Work (PoW) and standard cryptographic techniques are complicated, use too much energy, and could be vulnerable to future quantum computing attacks. Additionally, traditional cyberattack detection systems lack adaptability to evolving threats, thereby compromising the reliability and resilience of smart agriculture infrastructures. To overcome these limitations, this study proposes a quantum-secure, AI-empowered blockchain framework for smart agriculture. The framework integrates Post-Quantum Cryptography (PQC) using Crystals-Dilithium to ensure resistance against quantum attacks. A Delegated Proof of Stake (DPoS) consensus algorithm is implemented to enhance transaction validation speed and reduce energy consumption. Key generation processes are optimized using a hybrid metaheuristic approach that combines Whale Optimization Algorithm (WOA) with Particle Swarm Optimization (PSO). For finding unusual patterns, the system uses a Graph Neural Network (GNN) combined with TabNet, a type of deep learning model for tables, to effectively detect cyber threats in IoT environments. Experimental evaluation demonstrates that the proposed DPoS-based system validates transactions five times faster and consumes 40% less energy compared to the PoW mechanism. In terms of cybersecurity, the GNN–TabNet model achieves an accuracy of 97.8% and a false negative rate (FNR) of just 1.5%, outperforming conventional CNN-based models. Furthermore, the model exhibits an 18% reduction in memory usage, ensuring efficient operation in resource-constrained environments. This research presents a scalable, energy-efficient, and quantum-resistant framework tailored for smart agriculture.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

AI-optimized blockchain security for smart agriculture using post-quantum cryptography and graph neural network-based threat detection

  • 期刊论文

  • 发表日期：

    2025年9月1日

查看全文
  • Ibraheem M. Alharbi,
  • Najah Kalifah Almazmomi

摘要

The convergence of Artificial Intelligence (AI) and Blockchain technology has significantly advanced smart agriculture by enabling secure, transparent, and efficient data transactions. Even with these technological improvements, current blockchain security methods like Proof of Work (PoW) and standard cryptographic techniques are complicated, use too much energy, and could be vulnerable to future quantum computing attacks. Additionally, traditional cyberattack detection systems lack adaptability to evolving threats, thereby compromising the reliability and resilience of smart agriculture infrastructures. To overcome these limitations, this study proposes a quantum-secure, AI-empowered blockchain framework for smart agriculture. The framework integrates Post-Quantum Cryptography (PQC) using Crystals-Dilithium to ensure resistance against quantum attacks. A Delegated Proof of Stake (DPoS) consensus algorithm is implemented to enhance transaction validation speed and reduce energy consumption. Key generation processes are optimized using a hybrid metaheuristic approach that combines Whale Optimization Algorithm (WOA) with Particle Swarm Optimization (PSO). For finding unusual patterns, the system uses a Graph Neural Network (GNN) combined with TabNet, a type of deep learning model for tables, to effectively detect cyber threats in IoT environments. Experimental evaluation demonstrates that the proposed DPoS-based system validates transactions five times faster and consumes 40% less energy compared to the PoW mechanism. In terms of cybersecurity, the GNN–TabNet model achieves an accuracy of 97.8% and a false negative rate (FNR) of just 1.5%, outperforming conventional CNN-based models. Furthermore, the model exhibits an 18% reduction in memory usage, ensuring efficient operation in resource-constrained environments. This research presents a scalable, energy-efficient, and quantum-resistant framework tailored for smart agriculture.