Federated Learning is a distributed machine learning paradigm designed to protect data privacy and security while allowing multiple participants to collaboratively train a shared global model without uploading their data to a central server. Modern machine learning models often consist of millions of parameters, resulting in significant computational complexity and a heavy communication burden during the collection and distribution of training data. This becomes a key bottleneck that limits the effectiveness of federated learning. In order to reduce communication costs, we first propose the CS client selection strategy, which integrates model performance and fairness to construct scheduling priority metrics. Client selection is then carried out based on these metrics. And then, we propose a dynamic gradient sparsification algorithm that adjusts the level of sparsification applied to gradient updates for each client, ensuring that every client uploads a suitably compressed update. Finally, we conduct comprehensive experiments and the experimental results show that the method proposed in this paper can effectively reduce communication costs while ensuring model accuracy. © 2024 Copyright held by the owner/author(s).

This paper studies the difficult task of designing federated learning algorithms tailored for wireless ad hoc networks. Federated learning in such networks presents numerous challenges, including signal interference, decentralized infrastructure, dynamic changes in network topology, heterogeneous devices, and diverse data statistics. In response to these challenges, this paper proposes a fully distributed and personalized federated learning algorithm specifically designed for wireless ad hoc networks, named ADDPFed. ADDPFed tackles the issue of wireless interference by leveraging non-orthogonal multiple access technology and successive interference cancellation for enhancing overall communication efficiency. Another key strength of ADDPFed lies in its ability to enable direct local model exchanges among neighboring clients, eliminating the need for central server coordination for model aggregation. For the model aggregation, ADDPFed first calculates the distance between local models using the Tonimoto coefficient, and then assigns suited aggregation weights to these models. This approach emphasizes the contribution of similar models to the global model, effectively tackling the issue of data statistical heterogeneity. Extensive experiments validate the effectiveness of ADDPFed, paving the way for enhanced collaborative and distributed learning paradigms in wireless ad hoc networks. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Federated learning enables users to train models without transmitting local data to central servers, preserving data privacy and reducing communication costs. However, federated learning also faces various challenges, such as single point of failure issue, byzantine attack, and a lack of incentive mechanism. In this paper, a trustworthy and incentivized federated learning framework is proposed. Firstly, a blockchain network is used as a substitute for the central server in federated learning, addressing the vulnerability of servers to attacks and dishonesty. Secondly, a multi-key homomorphic encryption mechanism is introduced to safeguard the privacy of local models. Furthermore, to address the challenge of detecting malicious models within encrypted models, this paper proposes a verification mechanism based on the cosine similarity of group models. Finally, this paper evaluates the contributions of multiple participants using Shapley Value. This process effectively coordinates and distributes the interests of multiple participants, ultimately incenting more users to engage in federated learning. Experimental results demonstrate that the framework proposed in this paper can still converge and maintain good training accuracy even in the presence of attacks. © IFIP International Federation for Information Processing 2025.

Mobile edge computing aims to deploy mobile applications at the edge of wireless networks. Federated learning in mobile edge computing is a forward-looking distributed framework for deploying deep learning algorithms in many application scenarios. One challenge of federated learning in mobile edge computing is data heterogeneity since the unified model of federated learning performs poorly when client data are non-independent and identically distributed. Personalized federated learning can obtain amazing models in scenarios where client data are non-independent and identically distributed. This is because the personalized model captures the features of users' local data more accurately than the unified model. The personalized federated learning problem under two-tier (server-client) federated learning structures has been widely studied and applied. However, a lot of research results exhibit three distinct limitations: 1) suboptimal communication efficiency, 2) slow model convergence, and 3) underutilization of the relationships within user data, resulting in lower accuracy of personalized models. In this paper, we present the first personalized federated learning algorithm based on the client-edge-cloud structure. The edge server is responsible for model personalization and employs a learnable mixing parameter to mix the local model and the global model. We also utilize two learnable normalization parameters trained by clients to improve the performance of personalized models. Furthermore, in order to facilitate the collaboration among edge servers, we propose a similarity aggregation method to assign aggregation weights based on the Tanimoto coefficients between models. The experimental results show that the proposed algorithm not only increases the convergence speed of personalized models but also improves their testing accuracy.

As one of the essential steps to secure government data sharing, Identity Authentication (IA) plays a vital role in the processing of large data. However, the centralized IA scheme based on a trusted third party presents problems of information leakage and single point of failure, and those related to key escrow. Therefore, herein, an effective IA model based on multiattribute centers is designed. First, a private key of each attribute of a data requester is generated by the attribute authorization center. After obtaining the private key of attribute, the data requester generates a personal private key. Second, a dynamic key generation algorithm is proposed, which combines blockchain and smart contracts to periodically update the key of a data requester to prevent theft by external attackers, ensure the traceability of IA, and reduce the risk of privacy leakage. Third, the combination of blockchain and interplanetary file systems is used to store attribute field information of the data requester to further reduce the cost of blockchain information storage and improve the effectiveness of information storage. Experimental results show that the proposed model ensures the privacy and security of identity information and outperforms similar authentication models in terms of computational and communication costs.

Achieving efficient and secure sharing of data in the Internet of Vehicles (IoV) is of great significance for the development of smart transportation. Although blockchain technology has great potential to promote data sharing and privacy protection in the context of IoV, the problem of securing data sharing should be payed more attentions. This paper proposes an IoV data sharing scheme based on the hybrid architecture of blockchain and cloud-edge computing. Firstly, to improve protocol’s efficiency, a dual-chain structure empowered by alliance chain is introduced as the model architecture. Secondly, for the space problem characterized by data storage and security, we adopt distributed storage with the help of edge devices. Finally, to both ensure the efficiency of consensus protocol and protect the privacy of vehicles and owners simultaneously, we improve DPoS consensus algorithm to realize the efficient operation of the IoV data sharing model, which is closer to the actual needs of IoV. The comparison with other data sharing models highlights the advantages of this model, in terms of data storage and sharing security. It can be seen that the improved DPoS has high consensus efficiency and security in IoV. © 2023, The Author(s).

Public clouds have drawn increasing attention from academia and industry due to their high computational and storage performance. Attribute-based encryption (ABE) is the most promising technology to simultaneously achieve confidentiality and fine-grained access control of the cloud-stored data. However, traditional ABE that relies on centralized authority faces several key management issues, such as the key escrow, key distribution, key tracking, key update, and heavy communication and computing overhead for users, which will cause security concerns and impede its widespread application. On the other hand, blockchain technology preserves distributed ledgers to ensure the immutability and transparency of data, which can further solve the security vulnerabilities caused by system centralization. This paper proposes a blockchain-assisted transformation method to solve all the key management problems mentioned above in ciphertext-policy ABE by utilizing technologies such as secret sharing protocols. In addition, our transformation method realizes two additional benefits: outsourced decryption and efficient user revocation, which are extremely valuable for practical implementations. We simulate a demonstration by adopting the most popular permissioned blockchain, Hyperledger Fabric. The security and efficiency analysis reveals that the scheme obtained from our transformation method can achieve replayable chosen-ciphertext security with extremely efficient decryption.

Bitcoin uses a decentralized network of miners and a distributed consensus algorithm to agree on blockchains to process transactions, and designs certain incentive strategy to ensure the system run persistently. However, recent research finds that it is vulnerable to specific game-theoretic attacks, in which a rational attack can gain a disproportionate share of reward by deviating from the honest behaviors. Among these attacks, stubborn mining is generally regarded as the most effective one. This paper propose the optimal stubborn mining strategy, trying to obtain the maximum revenue in stubborn mining. Through careful analysis, we find that the mining strategies in stubborn mining, under different conditions, can be represented as a Markovian Decision Process (MDP), and solving the MDP can result in the optimal strategy. In solving the MDP, we first transform it into a transition-reward Matrices, and then evaluate it to get the largest reward. With the method mentioned above, the attackers can get 46.9% additional than honest miners, and this attack outweighs traditional selfish mining (a compelling and well-studied mining attack) by up to 7.53%. © 2023 Elsevier Inc.

Federated learning is an emerging paradigm that enables multiple organizations to jointly train a model without revealing their private data. As an important variant, vertical federated learning (VFL) deals with cases in which collaborating organizations own data of the same set of users but with disjoint features. It is generally regarded that VFL is more secure than horizontal federated learning. However, recent research (USENIX Security'22) reveals that it is still possible to conduct label inference attacks in VFL, in which attacker can acquire privately owned labels of other participants; even VFL constructed with model splitting (the kind of VFL architecture with higher security guarantee) cannot escape it. To solve this issue, in this paper, we propose the dispersed training framework. It utilizes secret sharing to break the correlations between the bottom model and the training data. Accordingly, even if the attacker receives the gradients in the training phase, he is incapable to deduce the feature representation of labels from the bottom model. Besides, we design a customized model aggregation method such that the shared model can be privately combined, and the linearity of secret sharing schemes ensures the training accuracy to be preserved. Theoretical and experimental analyses indicate the satisfactory performance and effectiveness of our framework.

Two packet scheduling algorithms for rechargeable sensor networks are proposed based on the signal to interference plus noise ratio model. They allocate different transmission slots to conflict-ing packets and overcome the challenges caused by the fact that the channel state changes quickly and is un-controllable. The first algorithm proposes a priority -based framework for packet scheduling in recharge-able sensor networks. Every packet is assigned a pri-ority related to the transmission delay and the remain-ing energy of rechargeable batteries, and the packets with higher priority are scheduled first. The second algorithm mainly focuses on the energy efficiency of batteries. The priorities are related to the transmis-sion distance of packets, and the packets with short transmission distance are scheduled first. The sensors are equipped with low-capacity rechargeable batteries, and the harvest-store-use model is used. We consider imperfect batteries. That is, the battery capacity is lim-ited, and battery energy leaks over time. The energy harvesting rate, energy retention rate and transmission power are known. Extensive simulation results indi-cate that the battery capacity has little effect on the packet scheduling delay. Therefore, the algorithms proposed in this paper are very suitable for wireless sensor networks with low-capacity batteries.