Collaborative Intrusion Detection in IoT by Integrating Blockchain and Semi-Supervised Learning

With the proliferation of Internet of Things (IoT) devices, traditional intrusion detection systems (IDS) face challenges such as widespread data distribution and high labeling costs. This paper proposes a blockchain-based semi-supervised collaborative intrusion detection system (B-IDS) that integrates blockchain technology with a semi-supervised Generative Adversarial Network (SGAN) model to enhance intrusion detection performance in IoT environments. B-IDS leverages the decentralized, tamper-proof, and traceable characteristics of blockchain to achieve secure data sharing and collaborative detection. Additionally, it utilizes InterPlanetary File System (IPFS) for large-scale data storage, alleviating the data transmission bottleneck associated with blockchain. Experimental results demonstrate that B-IDS outperforms traditional models in accuracy, recall, and F1 score, effectively enhancing the system's detection accuracy and resilience against attacks. This system exhibits high adaptability and application potential, offering a new solution for intrusion detection in IoT.