The blockchain-driven federated learning (FL) framework, leveraging its decentralized and immutable characteristics, enhances the security and credibility of IoT and edge computing. However, the inherent limitations in transaction throughput and high latency of the blockchain have posed substantial challenges to the efficiency of FL systems. Furthermore, FL itself is vulnerable to security threats such as poisoning attacks initiated by malicious participants and privacy inference attacks, which can compromise the security and confidentiality of the FL process. To address these challenges, this paper introduces a novel secure and trustworthy FL framework based on a hierarchical sharding blockchain structure, named HSChainFL. The design of HSChainFL framework centers around a meticulously crafted hierarchical sharding blockchain architecture. This structure facilitates interaction and collaboration between the main chain and multiple sub-chains, effectively reducing consensus overhead and thereby significantly improving the overall system efficiency. Moreover, HSChainFL incorporates a confidence evaluation method based on discrete cosine transform. This method utilizes the low-frequency components of the gradient vector to effectively identify malicious gradients in differential privacy scenarios, thereby providing strong resistance to poisoning attacks. Extensive experiments demonstrate that the HSChainFL framework exhibits superior performance in resisting various types of participant poisoning attacks while simultaneously reducing system overhead.

Public blockchain has outstanding performance in transaction privacy protection because of its anonymity. The data openness brings feasibility to transaction behavior analysis. At present, the transaction data of the public chain are huge, including complex trading objects and relationships. It is difficult to extract attributes and predict transaction behavior by traditional methods. To solve these problems, we extract transaction features to construct an Ethereum transaction heterogeneous information network (HIN) and propose a graph neural network (GNN)-based transaction prediction method for public blockchains in HINs, which can divide the network into subgraphs according to connectivity and increase the accuracy of the prediction results of transaction behavior. Experiments show that the execution time consumption of the proposed transaction subgraph division method is reduced by 70.61% on average compared with that of the search method. The accuracy of the proposed behavior prediction method also improves compared with that of the traditional random walk method, with an average accuracy of 83.82%.

The Bitcoin network comprises numerous nodes, necessitating users to invest significant network requests and time in comprehending its network topology. In this paper, we propose a Bitcoin network topology discovery algorithm that uses lightweight probe nodes to facilitate rapid transmission of network protocols. Building upon this, we introduce a node layer clustering algorithm based on filtering stable network nodes, enabling parallel discovery of the network topology. Additionally, we present an adaptive method for dynamically displaying the layered structure of the network topology. Experimental results demonstrate that our proposed method reduces communication overhead by approximately 72.16% when achieving a 95% similarity in network topology. Furthermore, the algorithm is applicable for discovering the network topology in other blockchain networks with similar structures.

In ocean environments, underwater sensor nodes (USNs) are susceptible to failure due to various factors, such as seawater corrosion, hardware failure, depleted battery, harsh deployment scenarios, and intentional sabotage. This article focuses on the topology restoration problem of disconnected subnetworks (TR-DSNs) caused by large-scale USN failures in underwater wireless sensor networks (UWSNs). The existing research cannot be well adapted to dynamic ocean environments because they ignore the effects of underwater communication channel and current movement on the cost and reliability of network restoration. It would consequently lead to high restoration cost and unreliable data transmission for UWSNs. To solve the mentioned problem, we first build a reliability evaluation model of topology restoration that considers the link quality, network connectivity, and data transmission of UWSNs in dynamic ocean environments. Then, a reliability-driven topology restoration strategy (called RDTRS) based on underwater relay node (URN) placement is designed. RDTRS comprises three key algorithms: 1) URN placement path generation; 2) URN location determination; and 3) URN location adjustment. By RDTRS, the number of URNs can be reduced on the premise of ensuring the restoration reliability of UWSNs. In the end, we validate the performance of RDTRS in terms of network restoration cost, packet delivery ratio, and transmission latency.

This paper proposes a novel Multiparty Oblivious Multivariate Polynomial Evaluation (MOMPE) protocol, which enables secure multivariate polynomial computation among multiple parties for the first time. Unlike traditional Oblivious Polynomial Evaluation (OPE) protocols limited to two-party secure computation, MOMPE divides participating parties into a sender holding the multivariate polynomial$P$and$n$receivers, each holding a secret value$\alpha_{i}$corresponding to a variable in$P$. Upon completion, the receivers learn$P\left(\alpha_{1}, \alpha_{2}, \ldots, \alpha_{n}\right)$, while the sender learns nothing. Assuming no collusion between the sender and receivers, we demonstrate MOMPE's security under the stringent Universal Composability (UC) framework, ensuring it remains secure in complex multiparty interactive environments. Our implementation and experiments validate MOMPE's performance, showing significantly less time consumption than the state-of-the-art Overdrive protocol for multiparty multivariate polynomial computation, while maintaining similar efficiency to OPE for participating parties. MOMPE has extensive application potential in privacy-preserving machine learning and collaborative data analysis, where multiple parties need to compute and analyze data together while protecting privacy. MOMPE offers a secure and efficient method for parties to utilize data without revealing sensitive information, opening new possibilities for privacy-preserving data collaboration and analysis.

The new generation of artificial intelligence (AI) technology will play an important role in promoting the digitalization, informatization and intelligence of the future power grid due to its high-dimensional state intelligent perception and rapid decision-making capabilities. However, its inherent shortcomings such as poor interpretability and fragility also limit the further application of AI technology in power systems. This paper first introduces hybrid-augmented intelligence (HAI) technology and its application development in the fields of autonomous driving and industrial robots. Combining the characteristics of the power system and AI technology, the requirements of the power system for HAI are analyzed and summarized. Secondly, the key technologies involved in human-machine collaborative HAI are analyzed in terms of data processing, model training and model application. On this basis, the application of HAI technology in typical scenarios such as power flow section regulation is designed and analyzed, which provides reference for subsequent engineering applications. Finally, the challenges faced by the application of HAI in power systems are analyzed and prospected, aiming to promote and enrich the development of basic theories and key technologies of hybrid intelligence in power systems.

The Bitcoin system uses anonymous transactions to protect users’ privacy, but attackers can use this defect of bitcoin transactions to discover the association between bitcoin addresses. At present, address clustering methods can make use of these vulnerabilities to associate the address as an entity to a certain extent. However, these address clustering methods have problems such as an insufficient inference rate of change addresses, inability to identify mixing transactions, and low efficiency of algorithm implementation. We propose some solutions to these problems. 1) We improve the method of change address identification to identify and mark more of them. 2) We propose a heuristic address clustering method related to mixing transactions, which can identify their privacy vulnerabilities. 3) We propose an incremental address clustering method that can store the historical state and more quickly discover the anonymity defect of Bitcoin. We use real Bitcoin transaction data to demonstrate our method’s feasibility and reliability.