This paper proposes a strategy for machine learning in the ciphertext domain.The data to be trained in the linear regression equation is encrypted by SHE homomorphic encryption,and then trained in the ciphertext domai...This paper proposes a strategy for machine learning in the ciphertext domain.The data to be trained in the linear regression equation is encrypted by SHE homomorphic encryption,and then trained in the ciphertext domain.At the same time,it is guaranteed that the error of the training results between the ciphertext domain and the plaintext domain is in a controllable range.After the training,the ciphertext can be decrypted and restored to the original plaintext training data.展开更多
In order to transmit the speech information safely in the channel,a new speech encryp-tion algorithm in linear canonical transform(LCT)domain based on dynamic modulation of chaot-ic system is proposed.The algorithm fi...In order to transmit the speech information safely in the channel,a new speech encryp-tion algorithm in linear canonical transform(LCT)domain based on dynamic modulation of chaot-ic system is proposed.The algorithm first uses a chaotic system to obtain the number of sampling points of the grouped encrypted signal.Then three chaotic systems are used to modulate the corres-ponding parameters of the LCT,and each group of transform parameters corresponds to a group of encrypted signals.Thus,each group of signals is transformed by LCT with different parameters.Fi-nally,chaotic encryption is performed on the LCT domain spectrum of each group of signals,to realize the overall encryption of the speech signal.The experimental results show that the proposed algorithm is extremely sensitive to the keys and has a larger key space.Compared with the original signal,the waveform and LCT domain spectrum of obtained encrypted signal are distributed more uniformly and have less correlation,which can realize the safe transmission of speech signals.展开更多
In this paper, Adomian decomposition method （ADM） with high accuracy and fast convergence is introduced to solve the fractional-order piecewise-linear （PWL） hyperchaotic system. Based on the obtained hyperchaotic ...In this paper, Adomian decomposition method （ADM） with high accuracy and fast convergence is introduced to solve the fractional-order piecewise-linear （PWL） hyperchaotic system. Based on the obtained hyperchaotic sequences, a novel color image encryption algorithm is proposed by employing a hybrid model of bidirectional circular permutation and DNA masking. In this scheme, the pixel positions of image are scrambled by circular permutation, and the pixel values are substituted by DNA sequence operations. In the DNA sequence operations, addition and substraction operations are performed according to traditional addition and subtraction in the binary, and two rounds of addition rules are used to encrypt the pixel values. The simulation results and security analysis show that the hyperchaotic map is suitable for image encryption, and the proposed encryption algorithm has good encryption effect and strong key sensitivity. It can resist brute-force attack, statistical attack, differential attack, known-plaintext, and chosen-plaintext attacks.展开更多
Although the learning with errors（LWE）-based full homomorphic encryption scheme was the first example of deviation from the original Gentry＇s blueprint, the scheme did not give detailed conversion process of circui...Although the learning with errors（LWE）-based full homomorphic encryption scheme was the first example of deviation from the original Gentry＇s blueprint, the scheme did not give detailed conversion process of circuit layer structure, and must rely on bootstrapping technique to achieve full homomorphism. Therefore, through modifying the re-linearization technique proposed by the above scheme, a technique called non-matrix key switching is presented, which includes key switching with re-linearization and pure key switching. The complex matrix operations of existing key switching technique are removed. Combining this technique with modulus switching, a （leveled） fully homomorphic encryption scheme without bootstrapping from LWE is constructed. In order to make circuit layer structure clear, the scheme gives detailed refresh door operation. Finally, we use bootstrapping to upgrade arithmetic circuit to any layer, and make the homomorphic computing capability of the scheme have nothing to circuit depth.展开更多
This paper observes approaches to algebraic analysis of GOST 28147-89 encryption algorithm (also known as simply GOST), which is the basis of most secure information systems in Russia. The general idea of algebraic an...This paper observes approaches to algebraic analysis of GOST 28147-89 encryption algorithm (also known as simply GOST), which is the basis of most secure information systems in Russia. The general idea of algebraic analysis is based on the representation of initial encryption algorithm as a system of multivariate quadratic equations, which define relations between a secret key and a cipher text. Extended linearization method is evaluated as a method for solving the nonlinear sys- tem of equations.展开更多
Any linear transform matrix can be used to easily calculate a consistent form, and a plurality of conversion can be easily connected together by matrix multiplication. When performing file transfers, you can encrypt f...Any linear transform matrix can be used to easily calculate a consistent form, and a plurality of conversion can be easily connected together by matrix multiplication. When performing file transfers, you can encrypt files matrix transformation. Article presents a matrix-based electronic document encryption and decryption algorithm, which relies on a special class of matrices combinatorial problems, the method to improve the security of electronic document system is feasible and effective, and finally give the source code and programming software.展开更多
Frequency Hopping Spread Spectrum (FHSS) system is often deployed to protect wireless communication from jamming or to preclude undesired reception of the signal. Such themes can only be achieved if the jammer or unde...Frequency Hopping Spread Spectrum (FHSS) system is often deployed to protect wireless communication from jamming or to preclude undesired reception of the signal. Such themes can only be achieved if the jammer or undesired receiver does not have the knowledge of the spreading code. For this reason, unencrypted M-sequences are a deficient choice for the spreading code when a high level of security is required. The primary objective of this paper is to analyze vulnerability of linear feedback shift register (LFSRs) codes. Then, a new method based on encryption algorithm applied over spreading codes, named hidden frequency hopping is proposed to improve the security of FHSS. The proposed encryption security algorithm is highly reliable, and can be applied to all existing data communication systems based on spread spectrum techniques. Since the multi-user detection is an inherent characteristic for FHSS, the multi-user interference must be studied carefully. Hence, a new method called optimum pair “key-input” selection is proposed which reduces interference below the desired constant threshold.展开更多
Based on analysis of the structure characteristics and implementation methods of some representative word oriented linear feedback shift registers （LFSRs） in several modem software oriented stream ciphers, this pape...Based on analysis of the structure characteristics and implementation methods of some representative word oriented linear feedback shift registers （LFSRs） in several modem software oriented stream ciphers, this paper firstly classifies the word oriented LFSRs into two classes： the machine instruction type and the arithmetic type. The similarities and differences between each type are illustrated by concrete examples. Then we give a detailed analysis about the word oriented LFSRs in each category from design structure, cryptographic properties and implementation issue aspects. Finally, some basic design criteria for modem word oriented LFSRs and suitable for software implementation are summarized.展开更多
A linearization attack on the Key Stream Generator （KSG） of the modified Eo algorithm proposed by Hermelin [Proceedings of ICISC＇99, Springer LNCS 1787, 2000, 17-29] is given in this paper. The initial value can be...A linearization attack on the Key Stream Generator （KSG） of the modified Eo algorithm proposed by Hermelin [Proceedings of ICISC＇99, Springer LNCS 1787, 2000, 17-29] is given in this paper. The initial value can be recovered by a linearization attack with O（2^60.52） operations by solving a System of Linear Equations （SLE） with at most 2^20.538 unknowns. Frederik Armknecht [Cryptology ePrint Archive, 2002/191] proposed a linearization attack on the KSG olEo algorithm with O（2^70.341） operations by solving an SLE with at most 2^24.056 unknowns, so the modification proposed by Hermelin reduces the ability or E0 to resist the linearization attack by comparing with the results ofFrederik Armknecht.展开更多
A new attack on block ciphers is introduced, which is termed linear-differential cryptanalysis. It bases the combining of linear cryptanalysis and differential cryptanalysis, and works by using linear-differential pro...A new attack on block ciphers is introduced, which is termed linear-differential cryptanalysis. It bases the combining of linear cryptanalysis and differential cryptanalysis, and works by using linear-differential probability （LDP）. Moreover, we present a new method for upper bounding the maximum linear-differential probability （MLDP） for 2 rounds of substitution permutation network （SPN） cipher structure. When our result applies to 2-round advanced encryption standard（AES）, It is shown that the upper bound of MLDP is up to 1.68×2^-19, which extends the known results for the 2-round SPN. Furthermore, when using a recursive technique, we obtain that the MLDP for 4 rounds of AES is bounded by 2^-73.展开更多
In the assessment of car insurance claims,the claim rate for car insurance presents a highly skewed probability distribution,which is typically modeled using Tweedie distribution.The traditional approach to obtaining ...In the assessment of car insurance claims,the claim rate for car insurance presents a highly skewed probability distribution,which is typically modeled using Tweedie distribution.The traditional approach to obtaining the Tweedie regression model involves training on a centralized dataset,when the data is provided by multiple parties,training a privacy-preserving Tweedie regression model without exchanging raw data becomes a challenge.To address this issue,this study introduces a novel vertical federated learning-based Tweedie regression algorithm for multi-party auto insurance rate setting in data silos.The algorithm can keep sensitive data locally and uses privacy-preserving techniques to achieve intersection operations between the two parties holding the data.After determining which entities are shared,the participants train the model locally using the shared entity data to obtain the local generalized linear model intermediate parameters.The homomorphic encryption algorithms are introduced to interact with and update the model intermediate parameters to collaboratively complete the joint training of the car insurance rate-setting model.Performance tests on two publicly available datasets show that the proposed federated Tweedie regression algorithm can effectively generate Tweedie regression models that leverage the value of data fromboth partieswithout exchanging data.The assessment results of the scheme approach those of the Tweedie regressionmodel learned fromcentralized data,and outperformthe Tweedie regressionmodel learned independently by a single party.展开更多
为解决医疗云平台共享个人健康档案(personal health record,PHR)存在的隐私泄露和加解密效率不理想的问题,以医疗云平台中帕金森病患者的转诊场景为例,提出了一种基于线性秘密共享的改进密文属性代理重加密方案(improved linear secret...为解决医疗云平台共享个人健康档案(personal health record,PHR)存在的隐私泄露和加解密效率不理想的问题,以医疗云平台中帕金森病患者的转诊场景为例,提出了一种基于线性秘密共享的改进密文属性代理重加密方案(improved linear secret sharing based ciphertext attribute proxy re-encryption scheme,LCPS)。该方案利用线性秘密共享技术来隐藏访问策略中的隐私属性,降低因访问策略暴露引发的隐私泄露风险;该方案还对代理重加密算法进行改进,通过减少复杂的双线性运算,提高了加解密效率。结果表明,LCPS在加解密方面的表现要优于其他方案。在判定性q-BDHE(q-decisional bilinear Diffie-Hellman exponent)困难假设下具有选择明文攻击时的不可区分性(indistinguishability under chosen-plaintext attack,IND-CPA)。该方案具有可移植性,同样适用于医疗云中其他病症转诊时的个人健康档案安全共享。展开更多
文摘This paper proposes a strategy for machine learning in the ciphertext domain.The data to be trained in the linear regression equation is encrypted by SHE homomorphic encryption,and then trained in the ciphertext domain.At the same time,it is guaranteed that the error of the training results between the ciphertext domain and the plaintext domain is in a controllable range.After the training,the ciphertext can be decrypted and restored to the original plaintext training data.
基金supported by the National Natural Science Found-ation of China(No.61901248)the Scientific and Tech-nological Innovation Programs of Higher Education Institu-tions in Shanxi(No.2019L0029).
文摘In order to transmit the speech information safely in the channel,a new speech encryp-tion algorithm in linear canonical transform(LCT)domain based on dynamic modulation of chaot-ic system is proposed.The algorithm first uses a chaotic system to obtain the number of sampling points of the grouped encrypted signal.Then three chaotic systems are used to modulate the corres-ponding parameters of the LCT,and each group of transform parameters corresponds to a group of encrypted signals.Thus,each group of signals is transformed by LCT with different parameters.Fi-nally,chaotic encryption is performed on the LCT domain spectrum of each group of signals,to realize the overall encryption of the speech signal.The experimental results show that the proposed algorithm is extremely sensitive to the keys and has a larger key space.Compared with the original signal,the waveform and LCT domain spectrum of obtained encrypted signal are distributed more uniformly and have less correlation,which can realize the safe transmission of speech signals.
基金supported by the National Natural Science Foundation of China(Grant Nos.61161006 and 61573383)
文摘In this paper, Adomian decomposition method （ADM） with high accuracy and fast convergence is introduced to solve the fractional-order piecewise-linear （PWL） hyperchaotic system. Based on the obtained hyperchaotic sequences, a novel color image encryption algorithm is proposed by employing a hybrid model of bidirectional circular permutation and DNA masking. In this scheme, the pixel positions of image are scrambled by circular permutation, and the pixel values are substituted by DNA sequence operations. In the DNA sequence operations, addition and substraction operations are performed according to traditional addition and subtraction in the binary, and two rounds of addition rules are used to encrypt the pixel values. The simulation results and security analysis show that the hyperchaotic map is suitable for image encryption, and the proposed encryption algorithm has good encryption effect and strong key sensitivity. It can resist brute-force attack, statistical attack, differential attack, known-plaintext, and chosen-plaintext attacks.
基金Supported by the National 863 Project(2012AA011705)Guangxi Natural Science Foundation(2013GXNSFBB053005)+2 种基金Guangxi Science Research&Technology Development Project(14124004-4-10)Guangdong Natural Science Foundation(2014A030313517)Guangxi Experiment Center of Information Science Foundation
文摘Although the learning with errors（LWE）-based full homomorphic encryption scheme was the first example of deviation from the original Gentry＇s blueprint, the scheme did not give detailed conversion process of circuit layer structure, and must rely on bootstrapping technique to achieve full homomorphism. Therefore, through modifying the re-linearization technique proposed by the above scheme, a technique called non-matrix key switching is presented, which includes key switching with re-linearization and pure key switching. The complex matrix operations of existing key switching technique are removed. Combining this technique with modulus switching, a （leveled） fully homomorphic encryption scheme without bootstrapping from LWE is constructed. In order to make circuit layer structure clear, the scheme gives detailed refresh door operation. Finally, we use bootstrapping to upgrade arithmetic circuit to any layer, and make the homomorphic computing capability of the scheme have nothing to circuit depth.
文摘This paper observes approaches to algebraic analysis of GOST 28147-89 encryption algorithm (also known as simply GOST), which is the basis of most secure information systems in Russia. The general idea of algebraic analysis is based on the representation of initial encryption algorithm as a system of multivariate quadratic equations, which define relations between a secret key and a cipher text. Extended linearization method is evaluated as a method for solving the nonlinear sys- tem of equations.
文摘Any linear transform matrix can be used to easily calculate a consistent form, and a plurality of conversion can be easily connected together by matrix multiplication. When performing file transfers, you can encrypt files matrix transformation. Article presents a matrix-based electronic document encryption and decryption algorithm, which relies on a special class of matrices combinatorial problems, the method to improve the security of electronic document system is feasible and effective, and finally give the source code and programming software.
文摘Frequency Hopping Spread Spectrum (FHSS) system is often deployed to protect wireless communication from jamming or to preclude undesired reception of the signal. Such themes can only be achieved if the jammer or undesired receiver does not have the knowledge of the spreading code. For this reason, unencrypted M-sequences are a deficient choice for the spreading code when a high level of security is required. The primary objective of this paper is to analyze vulnerability of linear feedback shift register (LFSRs) codes. Then, a new method based on encryption algorithm applied over spreading codes, named hidden frequency hopping is proposed to improve the security of FHSS. The proposed encryption security algorithm is highly reliable, and can be applied to all existing data communication systems based on spread spectrum techniques. Since the multi-user detection is an inherent characteristic for FHSS, the multi-user interference must be studied carefully. Hence, a new method called optimum pair “key-input” selection is proposed which reduces interference below the desired constant threshold.
基金Supported by the National Basic Research Program of China (937 Program) (2007CB807902)the National High-Technology Research and Development Program of China (863 Program) (2006AA01Z425)the National Natural Science Foundation of China (60503011, 90704003)
文摘Based on analysis of the structure characteristics and implementation methods of some representative word oriented linear feedback shift registers （LFSRs） in several modem software oriented stream ciphers, this paper firstly classifies the word oriented LFSRs into two classes： the machine instruction type and the arithmetic type. The similarities and differences between each type are illustrated by concrete examples. Then we give a detailed analysis about the word oriented LFSRs in each category from design structure, cryptographic properties and implementation issue aspects. Finally, some basic design criteria for modem word oriented LFSRs and suitable for software implementation are summarized.
文摘A linearization attack on the Key Stream Generator （KSG） of the modified Eo algorithm proposed by Hermelin [Proceedings of ICISC＇99, Springer LNCS 1787, 2000, 17-29] is given in this paper. The initial value can be recovered by a linearization attack with O（2^60.52） operations by solving a System of Linear Equations （SLE） with at most 2^20.538 unknowns. Frederik Armknecht [Cryptology ePrint Archive, 2002/191] proposed a linearization attack on the KSG olEo algorithm with O（2^70.341） operations by solving an SLE with at most 2^24.056 unknowns, so the modification proposed by Hermelin reduces the ability or E0 to resist the linearization attack by comparing with the results ofFrederik Armknecht.
基金Supported by the National Natural Science Foun-dation of China(60503010) and the Foundation of National Laboratory for Modern communications(51436030105DZ0105)
文摘A new attack on block ciphers is introduced, which is termed linear-differential cryptanalysis. It bases the combining of linear cryptanalysis and differential cryptanalysis, and works by using linear-differential probability （LDP）. Moreover, we present a new method for upper bounding the maximum linear-differential probability （MLDP） for 2 rounds of substitution permutation network （SPN） cipher structure. When our result applies to 2-round advanced encryption standard（AES）, It is shown that the upper bound of MLDP is up to 1.68×2^-19, which extends the known results for the 2-round SPN. Furthermore, when using a recursive technique, we obtain that the MLDP for 4 rounds of AES is bounded by 2^-73.
基金This research was funded by the National Natural Science Foundation of China(No.62272124)the National Key Research and Development Program of China(No.2022YFB2701401)+3 种基金Guizhou Province Science and Technology Plan Project(Grant Nos.Qiankehe Paltform Talent[2020]5017)The Research Project of Guizhou University for Talent Introduction(No.[2020]61)the Cultivation Project of Guizhou University(No.[2019]56)the Open Fund of Key Laboratory of Advanced Manufacturing Technology,Ministry of Education(GZUAMT2021KF[01]).
文摘In the assessment of car insurance claims,the claim rate for car insurance presents a highly skewed probability distribution,which is typically modeled using Tweedie distribution.The traditional approach to obtaining the Tweedie regression model involves training on a centralized dataset,when the data is provided by multiple parties,training a privacy-preserving Tweedie regression model without exchanging raw data becomes a challenge.To address this issue,this study introduces a novel vertical federated learning-based Tweedie regression algorithm for multi-party auto insurance rate setting in data silos.The algorithm can keep sensitive data locally and uses privacy-preserving techniques to achieve intersection operations between the two parties holding the data.After determining which entities are shared,the participants train the model locally using the shared entity data to obtain the local generalized linear model intermediate parameters.The homomorphic encryption algorithms are introduced to interact with and update the model intermediate parameters to collaboratively complete the joint training of the car insurance rate-setting model.Performance tests on two publicly available datasets show that the proposed federated Tweedie regression algorithm can effectively generate Tweedie regression models that leverage the value of data fromboth partieswithout exchanging data.The assessment results of the scheme approach those of the Tweedie regressionmodel learned fromcentralized data,and outperformthe Tweedie regressionmodel learned independently by a single party.
文摘为解决医疗云平台共享个人健康档案(personal health record,PHR)存在的隐私泄露和加解密效率不理想的问题,以医疗云平台中帕金森病患者的转诊场景为例,提出了一种基于线性秘密共享的改进密文属性代理重加密方案(improved linear secret sharing based ciphertext attribute proxy re-encryption scheme,LCPS)。该方案利用线性秘密共享技术来隐藏访问策略中的隐私属性,降低因访问策略暴露引发的隐私泄露风险;该方案还对代理重加密算法进行改进,通过减少复杂的双线性运算,提高了加解密效率。结果表明,LCPS在加解密方面的表现要优于其他方案。在判定性q-BDHE(q-decisional bilinear Diffie-Hellman exponent)困难假设下具有选择明文攻击时的不可区分性(indistinguishability under chosen-plaintext attack,IND-CPA)。该方案具有可移植性,同样适用于医疗云中其他病症转诊时的个人健康档案安全共享。
