The existing homomorphie eneryption scheme is based on ring of the integer, and the possible operators are restricted to addition and multiplication only. In this paper, a new operation is defined Similar Modul. Base ...The existing homomorphie eneryption scheme is based on ring of the integer, and the possible operators are restricted to addition and multiplication only. In this paper, a new operation is defined Similar Modul. Base on the Similar Modul, the number sets of the homomorphic encryption scheme is extended to real number, and the possible operators are extended to addition, subtraction, multiplication and division. Our new approach provides a practical ways of implementation because of the extension of the operators and the number sets.展开更多
Deep learning(DL)algorithms have been widely used in various security applications to enhance the performances of decision-based models.Malicious data added by an attacker can cause several security and privacy proble...Deep learning(DL)algorithms have been widely used in various security applications to enhance the performances of decision-based models.Malicious data added by an attacker can cause several security and privacy problems in the operation of DL models.The two most common active attacks are poisoning and evasion attacks,which can cause various problems,including wrong prediction and misclassification of decision-based models.Therefore,to design an efficient DL model,it is crucial to mitigate these attacks.In this regard,this study proposes a secure neural network(NN)model that provides data security during model training and testing phases.The main idea is to use cryptographic functions,such as hash function(SHA512)and homomorphic encryption(HE)scheme,to provide authenticity,integrity,and confidentiality of data.The performance of the proposed model is evaluated by experiments based on accuracy,precision,attack detection rate(ADR),and computational cost.The results show that the proposed model has achieved an accuracy of 98%,a precision of 0.97,and an ADR of 98%,even for a large number of attacks.Hence,the proposed model can be used to detect attacks and mitigate the attacker motives.The results also show that the computational cost of the proposed model does not increase with model complexity.展开更多
Privacy-preserving computational geometry is a special secure multi-party computation and has many applications. Previous protocols for determining whether a point is inside a circle are not secure enough. We present ...Privacy-preserving computational geometry is a special secure multi-party computation and has many applications. Previous protocols for determining whether a point is inside a circle are not secure enough. We present a two-round protocol for computing the distance between two private points and develop a more efficient protocol for the point-circle inclusion problem based on the distance protocol. In comparison with previous solutions, our protocol not only is more secure but also reduces the number of communication rounds and the number of modular multiplications significantly.展开更多
基金Supported by the National Natural Science Foun-dation of China (90104005)
文摘The existing homomorphie eneryption scheme is based on ring of the integer, and the possible operators are restricted to addition and multiplication only. In this paper, a new operation is defined Similar Modul. Base on the Similar Modul, the number sets of the homomorphic encryption scheme is extended to real number, and the possible operators are extended to addition, subtraction, multiplication and division. Our new approach provides a practical ways of implementation because of the extension of the operators and the number sets.
文摘Deep learning(DL)algorithms have been widely used in various security applications to enhance the performances of decision-based models.Malicious data added by an attacker can cause several security and privacy problems in the operation of DL models.The two most common active attacks are poisoning and evasion attacks,which can cause various problems,including wrong prediction and misclassification of decision-based models.Therefore,to design an efficient DL model,it is crucial to mitigate these attacks.In this regard,this study proposes a secure neural network(NN)model that provides data security during model training and testing phases.The main idea is to use cryptographic functions,such as hash function(SHA512)and homomorphic encryption(HE)scheme,to provide authenticity,integrity,and confidentiality of data.The performance of the proposed model is evaluated by experiments based on accuracy,precision,attack detection rate(ADR),and computational cost.The results show that the proposed model has achieved an accuracy of 98%,a precision of 0.97,and an ADR of 98%,even for a large number of attacks.Hence,the proposed model can be used to detect attacks and mitigate the attacker motives.The results also show that the computational cost of the proposed model does not increase with model complexity.
基金Supported by the National Natural Science Foundation of China （Grant No. 60573171）, the National Grand Fundaznental Research 973 Program of China, （Grant No. 2006CB303006）,and Research Program of Anhui Province Education Department （Grant Nos.2006KJ024A and JYXM2005166）. We are very grateful to Professor X. Yao at University of Birmingham for useful comments and some corrections. We also thank Professor H. Shen at Japhan Advanced Institute of Science and Technology for helpful suggestions.
文摘Privacy-preserving computational geometry is a special secure multi-party computation and has many applications. Previous protocols for determining whether a point is inside a circle are not secure enough. We present a two-round protocol for computing the distance between two private points and develop a more efficient protocol for the point-circle inclusion problem based on the distance protocol. In comparison with previous solutions, our protocol not only is more secure but also reduces the number of communication rounds and the number of modular multiplications significantly.
