No abstract

Current blockchain‐based energy trading models raise serious concerns regarding the high and capped transaction latency and expensive service charges. To solve these problems, In this article, we combine Lightning Network (LN) and smart contract to present a mobile energy trading scheme based on LN. The focal point of the scheme lies in transfer of value occurs off‐blockchain, which addresses the problem of transaction latency. Next, to solve the security of proposed scheme, we design a mechanism to delivery secret R. Participants exchange some parameters to calculate R instead of delivery it directly. The found security is guaranteed by committing funds into a multi‐signature address in proposed scheme. Then, we conduct a comprehensive experiment to evaluate the proposed scheme. The simulation and analysis results show that the proposed scheme is efficient than traditional trading. In addition, we analyze our entire scheme and conclude that it defenses anti‐tampering attacks and replay attack capacity effectively.

Multi-cloud computing provides services by used different clouds simultaneously multi-signature can be used as the interactive technology between multi-cloud and users. However, the limited resources of some terminal devices make multi-signature, which based on bilinear map, is not suitable for multi-cloud computing environment. In addition, the signers are disclosure in multi-signature so there is the risk of attack. To solve this issues, this paper proposes a certificateless designated verifier multi-signature scheme based on multivariable public key cryptography (MPKC). Firstly, the formalized definition and security model of the proposed scheme are given. Secondly, it is proved that the proposed scheme is against adaptive chosen-message attacks. Finally, the analysis shows that the proposed scheme is more efficiency and suitable for multi-cloud. Moreover, the proposed scheme can hidding signature source to achieve privacy protection.

Federated Learning (FL) is an emerging distributed framework that enables clients to conduct distributed learning and globally share models without requiring data to leave the local. In the FL process, participants are required to contribute data resources and computing resources for model training. However, the traditional FL lacks security guarantees and is vulnerable to attacks and damages by malicious adversaries. In addition, the existing incentive methods lack fairness to participants. Therefore, accurately identifying and preventing malicious nodes from doing evil, while effectively selecting and incentivizing participants, plays a vital role in improving the security and performance of FL.}{However, the existing FL lacks a malicious node detection mechanism, and also does not solve the cooperative fairness problem well. Therefore, correctly identifying and preventing malicious nodes from doing evil, effectively selecting and motivating high-quality participants to participate in training, are all key issues that we need to solve. In this paper, we propose a Robust Federated Learning Based on Malicious Detection and Incentives (MDIFL). Specifically, MDIFL first uses a gradient similarity to calculate reputation, thereby maintaining the reputation of participants and identifying malicious opponents, and then designs an effective incentive mechanism based on contract theory to achieve collaborative fairness. Extensive experimental results demonstrate that the proposed MDIFL can not only preferentially select and effectively motivate high-quality participants, but also correctly identify malicious adversaries, achieve fairness, and improve model performance.

No abstract