As an emerging joint learning model,federated learning is a promising way to combine model parameters of different users for training and inference without collecting users’original data.However,a practical and effic...As an emerging joint learning model,federated learning is a promising way to combine model parameters of different users for training and inference without collecting users’original data.However,a practical and efficient solution has not been established in previous work due to the absence of efficient matrix computation and cryptography schemes in the privacy-preserving federated learning model,especially in partially homomorphic cryptosystems.In this paper,we propose a Practical and Efficient Privacy-preserving Federated Learning(PEPFL)framework.First,we present a lifted distributed ElGamal cryptosystem for federated learning,which can solve the multi-key problem in federated learning.Secondly,we develop a Practical Partially Single Instruction Multiple Data(PSIMD)parallelism scheme that can encode a plaintext matrix into single plaintext for encryption,improving the encryption efficiency and reducing the communication cost in partially homomorphic cryptosystem.In addition,based on the Convolutional Neural Network(CNN)and the designed cryptosystem,a novel privacy-preserving federated learning framework is designed by using Momentum Gradient Descent(MGD).Finally,we evaluate the security and performance of PEPFL.The experiment results demonstrate that the scheme is practicable,effective,and secure with low communication and computation costs.展开更多
To address the scalability and identity federation problems of the traditional single sign-on system, the proposed scheme divides the security systems into different security domains. Each security domain has its own ...To address the scalability and identity federation problems of the traditional single sign-on system, the proposed scheme divides the security systems into different security domains. Each security domain has its own security servers and service providers, and there are trust relationships between different security domains for identity federation. The security server is responsible for authentication and authorization inside the domain, and offers identity federation capability for different domains. The security assertion markup language （SAML） assertion is used as security token in the system for authentication, authorization, and identity federation. The design of the proposed single sign-on process is based on web service security framework and multiple security domains, and the authorization is always deployed in the local area inside the service provider＇ s security domain, which enables web service clients, both inside and outside their security domains, to access the services in a simple, scalable, standard and secure way.展开更多
Single sign-on (SSO) is an identity management technique that provides the ability to use multiple Web services with one set of credentials. However, when the authentication server is down or unavailable, users cannot...Single sign-on (SSO) is an identity management technique that provides the ability to use multiple Web services with one set of credentials. However, when the authentication server is down or unavailable, users cannot access these Web services, regardless of whether they are operating normally. Therefore, it is important to enable continuous use alongside SSO. In this paper, we present an identity continuance method for SSO. First, we explain four such continuance methods and identify their limitations and problems. Second, we propose a new solution based on an identifier migration approach that meets the requirement for identity continuance. Finally, we discuss these methods from the viewpoint of continuity, security, efficiency, and feasibility.展开更多
基金supported by the National Natural Science Foundation of China under Grant No.U19B2021the Key Research and Development Program of Shaanxi under Grant No.2020ZDLGY08-04+1 种基金the Key Technologies R&D Program of He’nan Province under Grant No.212102210084the Innovation Scientists and Technicians Troop Construction Projects of Henan Province.
文摘As an emerging joint learning model,federated learning is a promising way to combine model parameters of different users for training and inference without collecting users’original data.However,a practical and efficient solution has not been established in previous work due to the absence of efficient matrix computation and cryptography schemes in the privacy-preserving federated learning model,especially in partially homomorphic cryptosystems.In this paper,we propose a Practical and Efficient Privacy-preserving Federated Learning(PEPFL)framework.First,we present a lifted distributed ElGamal cryptosystem for federated learning,which can solve the multi-key problem in federated learning.Secondly,we develop a Practical Partially Single Instruction Multiple Data(PSIMD)parallelism scheme that can encode a plaintext matrix into single plaintext for encryption,improving the encryption efficiency and reducing the communication cost in partially homomorphic cryptosystem.In addition,based on the Convolutional Neural Network(CNN)and the designed cryptosystem,a novel privacy-preserving federated learning framework is designed by using Momentum Gradient Descent(MGD).Finally,we evaluate the security and performance of PEPFL.The experiment results demonstrate that the scheme is practicable,effective,and secure with low communication and computation costs.
基金The National Natural Science Foundation of China(No60673054)
文摘To address the scalability and identity federation problems of the traditional single sign-on system, the proposed scheme divides the security systems into different security domains. Each security domain has its own security servers and service providers, and there are trust relationships between different security domains for identity federation. The security server is responsible for authentication and authorization inside the domain, and offers identity federation capability for different domains. The security assertion markup language （SAML） assertion is used as security token in the system for authentication, authorization, and identity federation. The design of the proposed single sign-on process is based on web service security framework and multiple security domains, and the authorization is always deployed in the local area inside the service provider＇ s security domain, which enables web service clients, both inside and outside their security domains, to access the services in a simple, scalable, standard and secure way.
文摘Single sign-on (SSO) is an identity management technique that provides the ability to use multiple Web services with one set of credentials. However, when the authentication server is down or unavailable, users cannot access these Web services, regardless of whether they are operating normally. Therefore, it is important to enable continuous use alongside SSO. In this paper, we present an identity continuance method for SSO. First, we explain four such continuance methods and identify their limitations and problems. Second, we propose a new solution based on an identifier migration approach that meets the requirement for identity continuance. Finally, we discuss these methods from the viewpoint of continuity, security, efficiency, and feasibility.
