With the flourishing development of philanthropy, more and more people are actively involved in charitable activities. However, traditional charitable systems face several challenges, including low levels of information, lack of transparency in the flow of funds, lack of accountability and vulnerability to tampering. To address these issues, Ethereum provides a distributed platform that can execute smart contracts, ensuring adherence to predefined rules and mitigating the possibility of fraud and third-party interference. Blockchain, as a decentralized distributed ledger technology, stores data records across multiple computer nodes and employs cryptographic methods to ensure data security and reliability. Its main features include decentralization, immutability, transparency and traceability. However, in the construction of a secure, reliable, and third-party-independent charitable donation system, challenges associated with privacy protection and search efficiency need to be addressed. To tackle these issues, this paper proposes a solution based on the Ethereum blockchain, highlighting the role of searchable encryption methods. This solution employs searchable encryption methods, enabling search operations to perform without decrypting the ciphertext, effectively safeguarding participants’ privacy information, and enhancing search efficiency. This system is more secure, trustworthy, and efficient compared to traditional charitable donation systems.In the implementation of the proposed solution, we conduct a detailed analysis of the security of the searchable encryption scheme. We successfully counter key guessing attacks and multi-keyword selection attacks. Furthermore, we perform a security analysis of the smart contract to verify the system’s feasibility and security. Additionally, we comprehensively compare the proposed solution with existing alternatives and evaluate its performance. By introducing Ethereum blockchain technology and searchable encryption methods, we provide a secure and reliable solution for the charitable donation field, significantly enhancing transparency and efficiency in charitable endeavors. This initiative not only safeguards the interests of donors and recipients but also makes a positive contribution to the development of social public welfare activities.

The Industrial Internet of Things (IIoT) is currently confronted with significant security challenges, including the complexities associated with cross-domain device permission management, the risks of privacy breaches during data transmission, and the vulnerabilities inherent in centralized security architectures to various forms of attack. Existing access control schemes are inadequate for addressing the dynamic and intricate nature of industrial environments, often falling short of meeting the IIoT system’s flexibility, security, and scalability requirements. This article introduces a novel hybrid access control scheme based on blockchain technology within a cross-domain context to tackle these issues. This approach synergizes the simplicity of role-based access control (RBAC) with the adaptability offered by attribute-based access control (ABAC) while harnessing blockchain’s decentralization, immutability, and transparency to bolster system security, autonomy, and capabilities for cross-domain management. We assessed this scheme’s effectiveness in defending against various threats through comprehensive security analysis. Furthermore, experimental results indicate that our proposed solution provides robust security guarantees and surpasses traditional approaches regarding efficiency and performance. Consequently, the scheme presented herein signifies a substantial innovation in IIoT security practices—offering a feasible and reliable solution tailored for complex cross-domain environments.

While transferring data to cloud servers frees users from having to manage it, it eventually raises new problems, such as data privacy. The concept of searchable encryption has drawn more and more focus in research as a means of resolving the tension between data accessibility and data privacy. Due to the lack of integrity and correctness authentication in most searchable encryption techniques, malicious cloud servers may deliver false search results to users. Based on public key encryption with searching (PEKS), the study suggests a privacy-preserving method for verifiable fuzzy keyword searches based on the Ethernet blockchain in a cloud context to overcome the aforementioned security concerns. The search user can check the accuracy and integrity of the query document using the unalterability characteristics of the Ethernet blockchain system in this scheme to prevent the cloud server from giving incorrect query results. Furthermore, a fair transaction between the cloud server and the data user is achieved and can be tracked back to the malicious user using hash functions and Ethereum smart contracts, even if the user or the cloud is malicious. Finally, the security analysis shows that, under the random oracle model, our technique fulfils the adaptive selection keyword’s semantic security. The performance assessment demonstrates that the proposed scheme outperforms other related schemes in terms of computational efficiency.

The continuous emergence of new attack classes challenges intrusion detection in edge Internet of Things (IoT) networks. Although federated learning enables distributed devices to collaboratively train a shared detector without exchanging raw traffic data, most federated intrusion detection systems assume a fixed label space. Retraining with all historical data incurs substantial storage and computation costs, whereas updating only with newly collected samples can cause catastrophic forgetting. The detector must mitigate catastrophic forgetting of previously observed attack classes while preserving sufficient new-class plasticity to learn emerging attacks under highly non-IID device data, intermittent client availability, constrained local memory, and repeated communication over bandwidth-limited and intermittently connected links. To address these challenges, this paper proposes EdgeFedCIL, a communication-efficient federated class-incremental intrusion detection framework. EdgeFedCIL preserves historical knowledge through client-local replay and knowledge distillation while reducing repeated model transmission through adaptive low-rank compression, quantization, and error feedback. A classifier-head protection strategy further limits compression-induced degradation of class discrimination. Experiments on public intrusion-detection datasets show that EdgeFedCIL achieves competitive or superior detection and historical-knowledge retention performance, particularly under highly heterogeneous client distributions, while reducing cumulative client-to-server model transmission by up to approximately 10.54 times relative to full-precision transmission. These results demonstrate the effectiveness of EdgeFedCIL for continual and communication-efficient intrusion detection in resource-constrained edge IoT networks.

The Industrial Internet of Things (IIoT) increasingly relies on federated learning (FL) to enable collaborative model training without directly sharing raw traffic data across industrial sites. However, in practical IIoT deployments, clients may later request the removal of their data contributions from a trained federated model due to regulatory requirements, such as the General Data Protection Regulation (GDPR), ownership transfer, or internal data-governance policies. Such practical requirements create a strong demand for federated unlearning in IIoT applications. Furthermore, IIoT deployments often exhibit highly imbalanced client data distributions, resulting in substantially different contributions of individual clients to the global model. Nevertheless, most existing federated unlearning methods adopt a uniform unlearning strategy and fail to account for such client-level contribution gaps. To address this issue, we propose DynamicFU, a contribution-aware dynamic federated unlearning framework for IIoT deployments. The proposed method evaluates the target client from parameter-level, data-level, and performance-level perspectives and adaptively determines the unlearning strength by dynamically adjusting the number of unlearning rounds. Experimental results on public IIoT datasets show that DynamicFU substantially improves unlearning efficiency, achieving up to 22.89× speedup over Full Retrain while maintaining comparable effectiveness.