Processing math: 1%
  • 中国精品科技期刊
  • CCF推荐A类中文期刊
  • 计算领域高质量科技期刊T1类
高级检索

面向机器学习的安全外包计算研究进展

陈珍珠, 周纯毅, 苏铓, 高艳松, 付安民

陈珍珠, 周纯毅, 苏铓, 高艳松, 付安民. 面向机器学习的安全外包计算研究进展[J]. 计算机研究与发展, 2023, 60(7): 1450-1466. DOI: 10.7544/issn1000-1239.202220767
引用本文: 陈珍珠, 周纯毅, 苏铓, 高艳松, 付安民. 面向机器学习的安全外包计算研究进展[J]. 计算机研究与发展, 2023, 60(7): 1450-1466. DOI: 10.7544/issn1000-1239.202220767
Chen Zhenzhu, Zhou Chunyi, Su Mang, Gao Yansong, Fu Anmin. Research Progress of Secure Outsourced Computing for Machine Learning[J]. Journal of Computer Research and Development, 2023, 60(7): 1450-1466. DOI: 10.7544/issn1000-1239.202220767
Citation: Chen Zhenzhu, Zhou Chunyi, Su Mang, Gao Yansong, Fu Anmin. Research Progress of Secure Outsourced Computing for Machine Learning[J]. Journal of Computer Research and Development, 2023, 60(7): 1450-1466. DOI: 10.7544/issn1000-1239.202220767
陈珍珠, 周纯毅, 苏铓, 高艳松, 付安民. 面向机器学习的安全外包计算研究进展[J]. 计算机研究与发展, 2023, 60(7): 1450-1466. CSTR: 32373.14.issn1000-1239.202220767
引用本文: 陈珍珠, 周纯毅, 苏铓, 高艳松, 付安民. 面向机器学习的安全外包计算研究进展[J]. 计算机研究与发展, 2023, 60(7): 1450-1466. CSTR: 32373.14.issn1000-1239.202220767
Chen Zhenzhu, Zhou Chunyi, Su Mang, Gao Yansong, Fu Anmin. Research Progress of Secure Outsourced Computing for Machine Learning[J]. Journal of Computer Research and Development, 2023, 60(7): 1450-1466. CSTR: 32373.14.issn1000-1239.202220767
Citation: Chen Zhenzhu, Zhou Chunyi, Su Mang, Gao Yansong, Fu Anmin. Research Progress of Secure Outsourced Computing for Machine Learning[J]. Journal of Computer Research and Development, 2023, 60(7): 1450-1466. CSTR: 32373.14.issn1000-1239.202220767

面向机器学习的安全外包计算研究进展

基金项目: 国家自然科学基金项目(62072239,62002167);江苏省自然科学基金项目(BK20211192,BK20200461);广西可信软件重点实验室研究课题(KX202029)
详细信息
    作者简介:

    陈珍珠: 1993年生. 博士研究生. 主要研究方向为云计算安全与机器学习安全

    周纯毅: 1995年生. 博士研究生. 主要研究方向为机器学习安全与隐私保护

    苏铓: 1987年生. 博士,副教授. 主要研究方向为安全访问控制与权限管理

    高艳松: 1986年生. 博士,副教授. 主要研究方向为硬件安全、人工智能安全和隐私、系统安全

    付安民: 1981年生. 博士,教授,博士生导师. CCF高级会员. 主要研究方向为密码学以及隐私保护

    通讯作者:

    付安民(fuam@njust.edu.cn

  • 中图分类号: TP391

Research Progress of Secure Outsourced Computing for Machine Learning

Funds: This work was supported by the National Natural Science Foundation of China(62072239, 62002167),the Natural Science Foundation of Jiangsu Province(BK20211192, BK20200461),and the Project of Guangxi Key Laboratory of Trusted Software(KX202029).
More Information
    Author Bio:

    Chen Zhenzhu: born in 1993. PhD candidate. Her main research interest includes cloud computing security and machine learning security

    Zhou Chunyi: born in 1995. PhD candidate. His main research interest includes machine learning security and privacy preserving

    Su Mang: born in 1987. PhD, associate professor. Her main research interests include secure access control and right management

    Gao Yansong: born in 1986. PhD, associate professor. His current research interests include hardware security, AI security and privacy, and system security

    Fu Anmin: born in 1981. PhD, professor, PhD supervisor. Senior member of CCF. His main research interests include cryptography and privacy preserving

  • 摘要:

    依靠机器学习,传统产业的数字化转型带来了海量数据增长,而产品服务的智能化提升则刺激了算力需求. 云计算的灵活资源调配可以为资源有限的企业和用户提供便宜便捷的外包计算服务,实现机器学习的模型训练和模型托管,加快产品和服务的智能化建设,促进数字经济增长. 然而,数据和模型外包伴随控制权转移,可能带来数据泄露风险和计算安全问题. 近年来,机器学习的外包安全问题受到越来越多研究者的关注,并取得了一些显著成果. 通过对2018—2022年这5年国内外机器学习安全外包研究工作调研,首先对现有主流的外包模型进行分类和特征归纳,依据任务阶段将外包模型划分为模型训练和模型托管模式,以及依据云服务商数量将外包模式划分为单云模式和多云模式. 其次重点从逻辑回归、朴素贝叶斯分类、支持向量机、决策树和神经网络等典型机器学习算法角度对机器学习安全外包计算相关研究进展进行了深入阐述和分析. 最后从不同角度分析和讨论了目前机器学习安全外包研究存在的不足,并展望未来面临的挑战和机遇.

    Abstract:

    Based on machine learning, the digital transformation of traditional industries brings a massive data growth, while the intelligent enhancement of products services raises the demand for computing power. Cloud computing, relying on flexible resource deployment, can provide inexpensive and convenient outsourced computing services for users with limited resources, enabling them to complete model training and model hosting for machine learning. It also contributes to the intelligent improvement of products and services and promotes the growth of the digital economy. However, data and model outsourcing come with a transfer of control, which may pose data leakage risk and computational security issues. In recent years, the security issues of machine learning outsourcing have received increasing public attentions and academic concerns. In this paper, we systematically reviewed the research work on machine learning security outsourcing in the year of 2018−2022 the past five years. We first present different outsourced modes, including model training and model hosting modes classified by the task phase, single-cloud and multi-cloud modes classified by the number of cloud service providers. Then we summarize the characteristics of outsourced models under different modes. Next, we focus on the research progress related to machine learning secure outsourced computing from the perspective of typical machine learning algorithms such as logistic regression, Bayesian classification, support vector machine, decision tree and neural network, and provide an in-depth description and analysis. Finally, we analyze and discuss the limitations from different perspectives, as well as potential challenges and opportunities.

  • 无线体域网[1](wireless body area network, WBAN)指由佩戴或嵌入在人体的各种无线传感器(wireless sensor, WS)组成的无线通信网络.WBAN技术在医疗数据监测方面的应用极为广泛,不同类型的无线医疗传感器负责监测患者各个方面的医疗数据并将数据发送给各种远端服务器,方便对患者的医疗数据做出专业的分析与整合.然而,开放的WBAN在传输患者敏感的医疗数据时,面临着患者的隐私被泄露或医疗数据被恶意篡改等风险[2].

    许多国内外学者提出将密码体制应用到WBAN中,以确保WBAN的医疗数据在传输与共享时的机密性.Mykletun等人[3]基于传统公钥密码(public key cryptography, PKC)体制,设计了一种保证无线传感网络数据机密性的加密方案.Nadir等人[4]基于PKC体制与椭圆曲线密码体制为用户生成对称密钥来加密数据,确保医疗数据在无线传感网络中传输与共享时的机密性.然而,基于PKC体制的方案[3-4]需要可信中心对用户证书进行管理,为消除证书管理的开销,一些基于身份加密体制的WBAN方案[5-7]相继被提出.上述文献[37]利用对数据进行加密的方式确保了医疗数据传输时的机密性,但这种方式没有实现对医疗数据来源的认证.如果无法实现医疗数据的可认证性,不仅会导致医院浪费宝贵的医疗资源进行无效的诊断,还可能基于被篡改的医疗数据而对患者的病情做出错误诊断.

    为了实现WBAN中医疗数据的可认证性,Ahn等人[8]构造了一种基于高级加密标准(advanced encryption standard,AES)对称密码体制的认证方案.黄一才等人[9]基于身份密码体制设计了一种签名方案,该方案实现了抗重放攻击.Cagalaban等人[10]将数字签密技术引入医疗保健系统,在确保医疗数据机密性的同时实现了数据的可认证性.Ullah等人[11]利用超椭圆曲线的概念,设计了一种基于证书的签密方案.尽管文献[811]实现了医疗数据的可认证性,但都没有考虑在多用户环境下的应用场景.为解决密码方案在多用户环境下的WBAN中计算效率较低的问题,基于聚合签名与聚合加密等技术,一些支持聚合模式的方案[12-15]相继被提出.然而,文献[815]没有考虑如何对WBAN云端密文进行有效的搜索,导致数据用户在对医疗数据进行检索时开销较大.

    基于可搜索加密技术[16]与密文等值测试技术[17],国内外学者提出了一些适用于WBAN的密文检索方案[18-21].但这些WBAN密文检索方案均存在一些缺陷,例如张嘉懿[18]与Andrew等人[19]提出的可搜索加密方案仅支持对用相同公钥加密的医疗数据进行搜索;Ramadan等人[20]设计的等值测试加密方案无法实现对医疗数据来源的认证;Elhabob等人[21]设计的基于证书的密文等值测试方案存在证书管理问题等.此外,医生或医疗机构有时需要判断多个患者某些特定方面的医疗数据是否相同,或对有相同病症的患者的医疗数据进行整合与存档,但密文检索文献[1821]均没有考虑到多用户检索以及对多密文同时进行检索的情况,在用户节点众多的WBAN实际应用环境中存在一定局限性.

    WBAN通常会面临需要对2个以上的密文进行匹配的情况,而传统的密文等值测试技术只能将多个密文两两分为一组,再对所有的分组逐个进行测试,在多用户环境下的密文检索效率较低.为提高密文等值测试技术在多密文测试时的计算效率,Susilo等人[22]提出了一种支持多密文等值测试的公钥加密(public-key encryption with multi-ciphertext equality test, PKE-MET)方案,实现了对2个以上的密文同时进行匹配的功能.在PKE-MET方案中,每个参与多密文等值测试的数据拥有者都可以指定1个数字n,并将自己的密文与其他n−1个数据拥有者的密文进行匹配.PKE-MET在支持同时对多密文进行等值测试的同时,还支持对多个用户同时进行密文检索,当测试者接收到n个希望进行密文检索的数据用户分别上传的n个测试陷门时,才可以对数据拥有者的密文进行测试,实现了多数据用户同时进行密文匹配的功能.然而,PKE-MET方案中存在证书管理开销较大、无法对数据的来源进行认证等问题.

    针对以上问题,本文提出了一种支持多密文等值测试的WBAN聚合签密方案.该方案的创新点主要包括3个方面:

    1)基于身份签密体制.本文方案采用基于身份的签密体制,消除了传统公钥加密方案中存在的证书管理开销,确保了WBAN中医疗数据的机密性、完整性、可认证性与数据拥有者签名的不可伪造性.

    2)支持多用户密文聚合签密.引入聚合签密技术,验证者可以实现对多个数据拥有者医疗数据密文的批量验证,提高了签密方案在多用户环境下的验证效率.

    3)支持多密文等值测试.引入多密文等值测试技术,测试者可以利用数据用户上传的测试陷门同时对多个密文进行匹配,实现了多用户检索与多密文等值测试,降低了多用户环境下等值测试过程的计算开销.

    计算性Diffie-Hellman(computation Diffie-Hellman, CDH)问题:给定(P,aP,bP),其中a,bZp,计算abP.

    由含有n个未知数x1,x2,,xnn个线性方程所组成的非齐次线性方程组

    {a11x1+a12x2++a1nxn=b1,a21x1+a22x2++a2nxn=b2, an1x1+an2x2++annxn=bn,

    所对应的系数矩阵为

    {\boldsymbol{A}} = \left({\begin{array}{*{20}{c}} {{a_{11}}}&{{a_{12}}}& \cdots &{{a_{1n}}} \\ {{a_{21}}}&{{a_{22}}}& \cdots &{{a_{2n}}} \\ \vdots & \vdots &{}& \vdots \\ {{a_{n1}}}&{{a_{n2}}}& \cdots &{{a_{nn}}} \end{array}} \right),

    矩阵A对应的行列式为

    \det ({\boldsymbol{A}}) = \left| {\begin{array}{*{20}{c}} {{a_{11}}}&{{a_{12}}}& \cdots &{{a_{1n}}} \\ {{a_{21}}}&{{a_{22}}}& \cdots &{{a_{2n}}} \\ \vdots & \vdots &{}& \vdots \\ {{a_{n1}}}&{{a_{n2}}}& \cdots &{{a_{nn}}} \end{array}} \right| \text{,}

    \det ({\boldsymbol{A}}) \ne 0,则该方程组有唯一解.

    形如

    {\boldsymbol{V}} = \left( {\begin{array}{*{20}{c}} 1&{{a_1}}&{a_1^2}& \cdots &{a_1^{n - 1}} \\ 1&{{a_2}}&{a_2^2}& \cdots &{a_2^{n - 1}} \\ \vdots & \vdots & \vdots &{}& \vdots \\ 1&{{a_n}}&{a_n^2}& \cdots &{a_n^{n - 1}} \end{array}} \right)

    的矩阵称为范德蒙矩阵,其对应的范德蒙行列式 \det ({\boldsymbol{V}}) 具有如下计算性质:

    \det ({\boldsymbol{V}}) = \left| {\begin{array}{*{20}{c}} 1&{{a_1}}&{a_1^2}& \cdots &{a_1^{n - 1}} \\ 1&{{a_2}}&{a_2^2}& \cdots &{a_2^{n - 1}} \\ \vdots & \vdots & \vdots &{}& \vdots \\ 1&{{a_n}}&{a_n^2}& \cdots &{a_n^{n - 1}} \end{array}} \right| = \prod\limits_{1 \leqslant i \lt j \leqslant n} {({a_i} - {a_j})} .

    本文提出的支持多密文等值测试的WBAN聚合签密方案的系统模型如图1所示,它包括6个实体:私钥生成器(private key generator, PKG)、云存储提供商、数据拥有者(即患者佩戴的无线传感器)、密文等值测试者、聚合者与数据用户(data user, DU).

    图  1  本文系统模型
    Figure  1.  The proposed system model

    各个实体具体介绍为:

    1)私钥生成器.负责为WBAN中的数据拥有者和数据用户生成密钥.

    2)云存储提供商.负责在云服务器中存储用户上传的医疗密文 C{T_1} C{T_2} ,…, C{T_n} .

    3)数据拥有者.即患者佩戴的无线传感器,负责对医疗数据进行签密并将医疗密文上传到云端存储.

    4)测试者.对从云服务器下载的多个医疗密文执行等值测试操作,将测试结果返回给云服务器.

    5)聚合者.负责对多个数据拥有者的医疗数据进行聚合签密,将聚合医疗密文上传到云端存储.

    6)数据用户.即医生、医疗机构与数据处理中心等希望获取医疗密文的用户,负责将等值测试的陷门上传给测试者,并对从云服务器下载的医疗密文进行解密与认证.

    本文提出的支持多密文等值测试的聚合签密方案需要考虑2种类型的敌手,第1类敌手无法访问数据用户的测试陷门,第2类敌手可以获取数据用户的测试陷门.针对这2类敌手,本文提出的方案旨在达到的安全目标为:

    1)医疗数据的机密性和完整性.WBAN中传输的大多是敏感的医疗数据,若患者的医疗数据在传输时中被恶意窃取或篡改,会造成严重后果.本文利用基于身份的加密体制,保证了所提方案在面对第1类攻击者时医疗数据的机密性与完整性.机密性指即使攻击者截取了传输的医疗密文也无法获取与明文相关的信息;完整性则指医疗数据在传输时中无法被敌手伪造或篡改.

    2)数据拥有者签名的不可伪造性.本文新方案在对数据拥有者的签名的合法性进行验证的过程中,采用基于身份的签密体制,保证了在面对第1类攻击者时数据拥有者签名的不可伪造性,即攻击者不能伪造出合法的数据拥有者签名.

    3)测试陷门的单向性.测试者通过数据用户上传的测试陷门对医疗密文进行等值测试操作,在测试过程中,需要保证面对第2类敌手时测试陷门满足单向性,即敌手无法通过测试陷门获取与参与测试的医疗数据明文相关的信息.

    给定安全参数 k ,PKG选择大素数 p ( p \gt {2^k} ), G 是阶为 p 的循环加法群, P G 的生成元.PKG随机选择 s \in \mathbb{Z}_p^* 作为主密钥秘密保存,计算 {P_{{\text{pub}}}} = sP 作为系统公钥,定义6个Hash函数: {H_1}:{\{ 0,1\} ^*} \to \mathbb{Z}_p^* {H_2}:{\{ 0,1\} ^*} \times G \to \mathbb{Z}_p^* {H_3}:{\{ 0,1\} ^*} \times G \to \mathbb{Z}_p^* {H_4}:G \to {\{ 0,1\} ^{{l_0} + {l_1}}} {H_5}:{\{ 0,1\} ^*} \to \mathbb{Z}_p^* {H_6}:{\{ 0,1\} ^*} \to {\{ 0,1\} ^k} ,其中 {l_0} 是密文长度.输出系统参数 params = \{ p,P,{P_{{\text{pub}}}},G,{H_1},{H_2},{H_3},{H_4},{H_5},{H_6}\} .

    1)用户将 I{D_i} 上传给PKG,PKG计算 {Q_i} = {H_1}(I{D_i}) s{k_{i,1}} = s{Q_i}

    2)PKG随机选择 {x_i} \in \mathbb{Z}_p^* ,计算 P{K_{i,1}}\; =\; {x_i}P P{K_{i,2}}\; = {H_1}(I{D_i}||P{K_{i,1}}) s{k_{i,2}} = {x_i} + sP{K_{i,2}} s{k_{i,3}} = {H_1}(I{D_i}||s) P{K_{i,3}} = s{k_{i,3}}P

    3)PKG输出公共参数 P{K_i} = (P{K_{i,1}},P{K_{i,2}},P{K_{i,3}}) 与私钥 s{k_i} = (s{k_{i,1}},s{k_{i,2}},s{k_{i,3}}) .

    给定参与密文等值测试与聚合签密的数据拥有者数量为 n ,数据拥有者的身份标识为 I{D_i} ,数据用户的身份标识为 I{D_j} ,其中i,j \in \{ 1,2, \cdots ,n\}.数据拥有者执行1)~5)操作对 {m_i} 进行签密:

    1)随机选择 {a_i},{b_i},{N_i} \in \mathbb{Z}_p^* ,计算 {C_{i,1}} = {a_i}P {C_{i,2}} = {b_i}P {R_i} = {a_i}{Q_j}{P_{{\text{pub}}}}

    2)计算 {U_i} = {H_2}({m_i},I{D_i},I{D_j},{R_i},P{K_{i,1}},P{K_{j,1}}) {V_i} = {H_3} ({m_i},I{D_i},I{D_j},{R_i},P{K_{i,1}},P{K_{j,1}}) {v_i} = {a_i}{U_i} + s{k_{i,2}}{V_i} {C_{i,3}} = {v_i}P {C_{i,4}} = {H_4}({R_i}) \oplus ({m_i}||{v_i})

    3)计算 {f_{i,0}} = {H_5}({m_i}||n) {f}_{i,1} = {H}_{5}({m}_{i}|\left|n\right||{f}_{i,0}),\cdots {f_{i,n - 1}} = {H_5}({m_i}||n||{f_{i,0}}|| \cdots ||{f_{i,n - 2}})

    4)计算 {C_{i,5}} \;= \;{H_4}({b_i}P{K_{j,3}}) \;\oplus\; ({N_i}||f({N_i})){C_{i,6}}\; = \;{H_6} (n|| {C_{i,1}}|| \cdots ||{C_{i,5}}||{b_i}P{K_{j,3}}||{f_{i,0}}|| \cdots ||{f_{i,n - 1}}),其中 f({N_i}) = {f_{i,0}} + {f_{i,1}}{N_i} + {f_{i,2}}N_i^2 + \cdots + {f_{i,n - 1}}N_i^{n - 1}

    5)将密文 C{T_i} = ({t_i},{C_{i,1}},{C_{i,2}},{C_{i,3}},{C_{i,4}},{C_{i,5}},{C_{i,6}}) 上传到云端存储,其中 {t_i} = n .

    n 个数据用户分别将等值测试陷门 t{k_j} = s{k_{j,3}} 发送给测试者,其中j \in \{ 1,2, \cdots ,n\}.测试者从云服务器分别下载 n 个数据拥有者想要测试的密文 C{T_1,CT_2,\cdots,CT_n} ,执行1)~3)多密文等值测试操作:

    1)检查{t_1} = {t_2} = \cdots = {t_n} = n是否成立,若成立测试者则继续执行以下操作,否则终止操作并输出“ \bot ”;

    2)对于 i \in \{ 1,2, \cdots ,n\} j \in \{ 1,2, \cdots ,n\} ,测试者分别计算 {N_i}||f({N_i}) = {C_{i,5}} \oplus {H_4}({C_{i,2}}t{k_j}) ,由签密算法有 f({N_i}) = {f_{i,0}} + {f_{i,1}}{N_i} + {f_{i,2}}N_i^2 + \cdots + {f_{i,n - 1}}N_i^{n - 1} ,测试者将 n 个等式合并得到方程组

    \left\{\begin{aligned} &f({N}_{1})={f}_{1,0}+{f}_{1,1}{N}_{1}+{f}_{1,2}{N}_{1}^{2}+\cdots +{f}_{1,n-1}{N}_{1}^{n-1},\\ &f({N}_{2})={f}_{2,0}+{f}_{2,1}{N}_{2}+{f}_{2,2}{N}_{2}^{2}+\cdots +{f}_{2,n-1}{N}_{2}^{n-1},\\ & \;\;\; \vdots \\ &f({N}_{n})={f}_{n,0}+{f}_{n,1}{N}_{n}+{f}_{n,2}{N}_{n}^{2}+\cdots +{f}_{n,n-1}{N}_{n}^{n-1},\end{aligned}\right.

    并隐式设置 {f_{i,k}} = {f_{j,k}} ,其中 k \in \{ 0,1, \cdots ,n - 1\} ,测试者通过对该方程组对应的范德蒙矩阵求逆,获得方程组的唯一一组解 {f_{1,0}},{f_{1,1}}, \cdots ,{f_{1,n - 1}}

    3)检查等式{C_{i,6}} = {H_6}(n||{C_{i,1}}||{C_{i,2}}||{C_{i,3}}||{C_{i,4}}||{C_{i,5}}||{C_{i,2}}t{k_j}|| {f_{i,0}}||{f_{i,1}}|| \cdots ||{f_{i,n - 1}})是否成立,若成立测试者则向云服务器输出测试结果为“1”,否则向云服务器输出测试结果为“0”.

    若云服务器接收到的密文等值测试结果为“1”,代表 n 个数据拥有者的医疗密文全部相同,云服务器将所有数据拥有者的医疗密文 C{T}_{1},C{T}_{2},\cdots ,C{T}_{n} 发送给聚合者,聚合者执行1)~2)操作对医疗密文进行聚合签密:

    1)计算{X_{{\text{agg}}}} = \displaystyle\sum\limits_{i = 1}^n {{C_{i,3}}}

    2)将聚合医疗密文 {\sigma _{{\text{agg}}}} = ({\{ C{T_i}\} _{i = 1,2, \cdots ,n}},{X_{{\text{agg}}}}) 上传到云服务器存储.

    给定数据用户的身份标识为 I{D_j} ,其中 j \in \{ 1, 2, \cdots , n\} .数据用户从云端下载聚合医疗密文 {\sigma _{{\text{agg}}}} ,对密文进行解密并验证数据来源.数据用户的具体操作如为:

    1)计算R_{i}'= sk_{j,1} C_{i,1}m_i'||v_i' = {C_{i,4}} \oplus {H_4}(R_i')

    2)根据m_i'的值计算{f}_{i,0}'\;=\;{H}_{5}({m}_{i}'||n),f_{i,1}^{{'} }\; =\; {H_5}(m_i^{{'} }||n|| f_{i,0}^{{'} }) ,\cdotsf_{i,n - 1}^{'} = {H_5}(m_i'||n||f_{i,0}'||, \cdots ||f_{i,n - 2}^{{'} })N_i^{{'} }||f(N_i^{{'} }) = {C_{i,5}} \oplus {H_4} ({C_{i,2}}s{k_{j,3}})

    3)计算U_i^{{'} } = {H_2}(m_i^{{'} },I{D_i},I{D_j},R_i^{{'} },P{K_{i,1}},P{K_{j,1}})V_i' = {H_3} (m_i', \; I{D_i},\;I{D_j},\;R_i',\;P{K_{i,1}},\;P{K_{j,1}})X_{{\text{agg}}}' = \displaystyle\sum\limits_{i = 1}^n {v_i'P}X_{{\text{agg}}}^*= \displaystyle\sum\limits_{i = 1}^n {U_i'{C_{i,1}} +} \displaystyle\sum\limits_{i = 1}^n {V_i'P{K_{i,1}} + }\displaystyle\sum\limits_{i = 1}^n {V_i'P{K_{i,2}}{P_{{\text{pub}}}}}

    4)分别检查等式{C_{i,6}}\; =\; {H_6}(n||{C_{i,1}}||{C_{i,2}}||{C_{i,3}}||{C_{i,4}}||{C_{i,5}}|| {C_{i,2}}s{k_{j,3}}|| f_{i,0}'||f_{i,1}'|| \cdots ||f_{i,n - 1}')X_{{\text{agg}}}^* = X_{{\text{agg}}}'f(N_i') = f_{i,0}' + {f_{i,1}'N_i'} +\cdots+ f_{i,n-1}'N_i^{{'}n-1}是否同时成立.

    若以上等式均成立,数据用户则接收医疗数据m_i';否则输出“ \bot ”.

    1)解密等式的正确性

    数据用户通过计算 m_i'||v_i' = {C_{i,4}} \oplus {H_4}(R_i') 对密文进行解密,其中 R_i' = s{k_{j,1}}{C_{i,1}} s{k_{j,1}} 是数据用户的私钥,由于s{k_{j,1}} = s{Q_j},则有

    R_i' = s{k_{j,1}}{C_{i,1}} = s{k_{j,1}}{a_i}P = s{Q_j}{a_i}P = {a_i}{Q_j}{P_{{\text{pub}}}} = {R_i} \text{,}

    R_i' = {R_i},从而有

    m_i'||v_i' = {C_{i,4}} \oplus {H_4}(R_i') = {H_4}({R_i}) \oplus ({m_i}||{v_i}) \oplus {H_4}(R_i') = {m_i}||{v_i}{\kern 1pt} .

    因此,本文方案满足密文解密等式的正确性.

    2)签名验证等式的正确性

    数据用户通过判断等式X_{{\text{agg}}}^* = X_{{\text{agg}}}'是否成立以验证聚合密文签名的合法性,其中X_{{\text{agg}}}' = \displaystyle\sum\limits_{i = 1}^n {v_i'P}{v_i'} = {a_i}{U_i} +s{k_{i,2}}{V_i} s{k_{i,2}} = {x_i} + sP{K_{i,2}} ,则有

    \begin{aligned} X_{{\text{agg}}}' = &\sum\limits_{i = 1}^n {v_i'P} = \sum\limits_{i = 1}^n {{a_i}{U_i}P + \sum\limits_{i = 1}^n {s{k_{i,2}}{V_i}P} } = \\ &\sum\limits_{i = 1}^n {{a_i}{U_i}P + \sum\limits_{i = 1}^n {{x_i}{V_i}P + \sum\limits_{i = 1}^n {sP{K_{i,2}}{V_i}P} } } ,\end{aligned}

    结合 {C_{i,1}} = {a_i}P P{K_{i,1}} = {x_i}P {P_{{\text{pub}}}} = sP ,从而有

    X_{{\text{agg}}}' = \sum\limits_{i = 1}^n {{U_i}{C_{i,1}} + } \sum\limits_{i = 1}^n {{V_i}P{K_{i,1}} + } \sum\limits_{i = 1}^n {{V_i}P{K_{i,2}}{P_{{\text{pub}}}}}.

    进一步,由解密等式的正确性可知 m_i'||v_i' = {m_i}||{v_i} ,则有

    \begin{aligned} {U_i} =\;& {H_2}({m_i},I{D_i},I{D_j},{R_i},P{K_{i,1}},P{K_{j,1}})= \\ & {H_2}(m_i',I{D_i},I{D_j},R_i',P{K_{i,1}},P{K_{j,1}}) =U_i',\\ {V_i} = & {H_3}({m_i},I{D_i},I{D_j},{R_i},P{K_{i,1}},P{K_{j,1}}) =\\ &{H_3}(m_i',I{D_i},I{D_j},R_i',P{K_{i,1}},P{K_{j,1}}) = V_i', \end{aligned}

    {U_i} = U_i' {V_i} = V_i' ,于是有

    \begin{aligned} X_{{\text{agg}}}' = \;& \sum\limits_{i = 1}^n {{U_i}{C_{i,1}} + } \sum\limits_{i = 1}^n {{V_i}P{K_{i,1}} + } \sum\limits_{i = 1}^n {{V_i}P{K_{i,2}}{P_{{\text{pub}}}}} = \\ &\sum\limits_{i = 1}^n {U_i^{'}{C_{i,1}} + } \sum\limits_{i = 1}^n {V_i'P{K_{i,1}} + } \sum\limits_{i = 1}^n {V_i'P{K_{i,2}}{P_{{\text{pub}}}}} = X_{{\text{agg}}}^* \text{,} \end{aligned}

    X_{{\text{agg}}}^* = X_{{\text{agg}}}' 成立.因此,本文所提的新方案满足签名验证等式的正确性.

    3)等值测试结果的正确性

    i \in \{ 1,2, \cdots ,n\} j \in \{ 1,2, \cdots ,n\} ,测试者通过检查 {C_{i,6}} = {H_6}(n||{C_{i,1}}|| \cdots ||{C_{i,5}}||{C_{i,2}}t{k_j}||{f_{i,0}}|| \cdots ||{f_{i,n - 1}}) 是否成立来判断 n 个医疗密文是否相同,其中{f_{i,0}}\; =\; {H_5} ({m_i}|| n), \cdots , {f_{i,n - 1}} = {H_5}({m_i}||n||{f_{i,0}}|| \cdots ||{f_{i,n - 2}}) .假设所有参与密文等值测试的医疗密文全部相同,即 {m_1} = {m_2} = \cdots = {m_n} ,则有

    \begin{aligned} {H}_{5}({m}_{1}||n)={H}_{5}({m}_{2}||n)=\; &\cdots ={H}_{5}({m}_{n}||n),\\ {H}_{5}({m}_{1}|\left|n\right||{f}_{1,0})={H}_{5}({m}_{2}|\left|n\right|| & {f}_{1,0})= \cdots ={H}_{5}({m}_{n}|\left|n\right||{f}_{1,0}),\\ &\vdots\\ {H}_{5}({m}_{1}||n||{f}_{1,0}||\cdots ||{f}_{1,n-2})= & {H}_{5}({m}_{1}||n||{f}_{2,0}||\cdots ||{f}_{2,n-2})=\cdots=\\ {H}_{5}({m}_{n}||n||{f}_{n,0}||&\cdots ||{f}_{n,n-2}), \end{aligned}

    即对于所有的 i,j \in \{ 1,2, \cdots ,n\} k \in \{ 0,1, \cdots ,n - 1\} ,等式 {f_{i,k}} = {f_{j,k}} 均成立.

    由医疗数据签密及上传算法可知,数据拥有者在签密过程中设置

    f({N_i}) = {f_{i,0}} + {f_{i,1}}{N_i} + {f_{i,2}}N_i^2 + \cdots + {f_{i,n - 1}}N_i^{n - 1},

    由此可以得到方程组

    \left\{\begin{aligned} f({N}_{1})&={f}_{1,0}+{f}_{1,1}{N}_{1}+{f}_{1,2}{N}_{1}^{2}+\cdots +{f}_{1,n-1}{N}_{1}^{n-1},\\ f({N}_{2})&={f}_{2,0}+{f}_{2,1}{N}_{2}+{f}_{2,2}{N}_{2}^{2}+\cdots +{f}_{2,n-1}{N}_{2}^{n-1},\\ & \vdots \\ f({N}_{n})&={f}_{n,0}+{f}_{n,1}{N}_{n}+{f}_{n,2}{N}_{n}^{2}+\cdots +{f}_{n,n-1}{N}_{n}^{n-1},\end{aligned}\right.

    结合 {f_{i,k}} = {f_{j,k}} ,因此可将 {f_{1,0}},{f_{1,1}}, \cdots ,{f_{1,n - 1}} 作为方程组的解,将随机数 {N_i} 作为方程组的系数,则该方程组对应的矩阵为

    {\boldsymbol{V}} = \left({\begin{array}{*{20}{c}} 1&{{N_1}}&{N_1^2}& \cdots &{N_1^{n - 1}} \\ 1&{{N_2}}&{N_2^2}& \cdots &{N_2^{n - 1}} \\ \vdots & \vdots & \vdots &{}& \vdots \\ 1&{{N_n}}&{N_n^2}& \cdots &{N_n^{n - 1}} \end{array}} \right) ,

    由范德蒙矩阵的性质可知其对应的行列式为 \det ({\boldsymbol{V}}) = \displaystyle\prod\limits_{1 \leqslant i \lt j \leqslant n} {({N_i} - {N_j})} .

    从数据拥有者签密过程可知, {N_i} 是由 n 个不同的数据拥有者在对医疗密文进行签密时分别选择的随机数,因此 \det ({\boldsymbol{V}}) = 0 的概率仅为 {[p(p - 1) \cdots (p - n + 1)]^{ - 1}} ,其中 p 为群 \mathbb{Z}_p^* 的阶.由克拉默法则可知当 \det ({\boldsymbol{V}}) \ne 0 时,方程组有且仅有唯一解 {f_{1,0}},{f_{1,1}}, \cdots ,{f_{1,n - 1}} ,于是有对于所有的 i,j \in \{ 1,2, \cdots ,n\} k \in \{ 0,1, \cdots ,n - 1\} ,等式 {f_{i,k}} = {f_{j,k}} 均成立,与所有参与密文等值测试的医疗密文全部相同的假设相符.因此,本文新方案满足多密文等值测试结果的正确性.

    本文提出的方案引入了基于身份的聚合签密体制,确保了本文方案在面对第1类敌手时医疗数据的机密性与签名的存在不可伪造性,对于机密性与不可伪造性的证明过程可以参考文献[23]方案.同时,本文方案满足面对第2类敌手适应性选择密文攻击下的单向性(one-way against adaptive chosen ciphertext attack, OW-CCA2),以下通过定理1证明本文方案满足OW-CCA2安全.

    定理1. 假设CDH问题是难解的,则本文方案在随机预言模型下对第2类敌手是OW-CCA2安全的.

    证明.假设 \mathcal{C} 是能够解决CDH困难问题的人, {\mathcal{A}_2} 代表第2类敌手. \mathcal{C} {\mathcal{A}_2} 为子程序充当以下游戏中的挑战者,若 {\mathcal{A}_2} 能以不可忽略的优势在概率多项式时间内的游戏中获胜,则 \mathcal{C} 能够在概率多项式时间内解决CDH困难问题.

    初始化阶段.CDH问题的输入为 (P,aP,bP) ,其中 a,b \in \mathbb{Z}_p^* \mathcal{C} 的目标是给出CDH困难问题的解 abP . \mathcal{C} 选取阶为素数 p 的循环群 G ,计算 P G 的生成元,随机选择 a \in \mathbb{Z}_p^* 并计算P_{{\text{pub}}}' = aP.最后,输出系统参数 params=\{p,P,{P}_{\text{pub}},G,{H}_{1},{H}_{2},{H}_{3},{H}_{4},{H}_{5},{H}_{6}\} ,将 a 秘密保存并发送 params {\mathcal{A}_2} .

    询问阶段1.为了响应 {\mathcal{A}_2} 的询问, \mathcal{C} 维持列表 {L}_{1}, {L}_{2},{L}_{3},{L}_{4},{L}_{5},{L}_{6},{L}_{\text{td}} 分别用于跟踪 {\mathcal{A}_2} {H_1} Hash询问、 {H_2} Hash询问、 {H_3} Hash询问、 {H_4} Hash询问、 {H_5} Hash询问、 {H_6} Hash询问、测试陷门询问. {L_1} 同时用于跟踪密钥提取询问,开始时每个列表都为空.

    1) {H_1} Hash询问.当 \mathcal{C} 收到 {\mathcal{A}_2} {H_1}(I{D_i},{Q_i}) 的查询,若 I{D_i} \in \{ I{D_i}\} _{i = 1}^n ,则计算 P{K_{i,1}} = {x_i}P ,其中 {x_i} 是未知的, \mathcal{C} 保存 ( \bot ,{Q_i},I{D_i}) {L_1} ;若 i \ne 1 \mathcal{C} 随机选择 {x_i},P{K_{i,2}} \in \mathbb{Z}_p^* 并设置 P{K_{i,1}} = {x_i}P ,将 P{K_{i,2}} = {H_1}(I{D_i}||P{K_{i,1}}) 返回给 {\mathcal{A}_2} 并保存 ({x_i},P{K_{i,1}},P{K_{i,2}},I{D_i}) {L_1} .

    2) {H_2} Hash询问.当 \mathcal{C} 收到 {\mathcal{A}_2} ({m_i},I{D_i},I{D_j},{R_i}, P{K_{i,1}},P{K_{j,1}},{U_i})的查询后, \mathcal{C} 首先在 {L_2} 查找是否已有({m_i}, I{D_i},I{D_j},{R_i},P{K_{i,1}},P{K_{j,1}},{U_i},{t_i},{t_i}P),若 {L_2} 已有({m_i},I{D_i}, I{D_j},{R_i},P{K_{i,1}},P{K_{j,1}},{U_i},{t_i},{t_i}P),则发送 {U_i} {\mathcal{A}_2} ;否则, \mathcal{C} 选取 {U_i} \in \mathbb{Z}_p^* ,将 ({U_i},{t_i},{t_i}P) 加入到 {L_2} 中并输出 {t_i}P .

    3) {H_3} Hash询问.当 \mathcal{C} 收到 {\mathcal{A}_2} ({m_i},I{D_i},I{D_j},{R_i}, P{K_{i,1}}, P{K_{j,1}},{V_i})的查询后, \mathcal{C} 首先在 {L_3} 查找是否已有({m_i}, I{D_i}, I{D_j},{R_i},P{K_{i,1}},P{K_{j,1}},{V_i},{w_i},{w_i}P),若 {L_3} 已有({m_i},I{D_i}, I{D_j},{R_i},P{K_{i,1}},P{K_{j,1}},{V_i},{w_i},{w_i}P),则返回 {V_i} {\mathcal{A}_2} ;否则, \mathcal{C} 选取 {V_i} \in \mathbb{Z}_p^* ,将 ({V_i},{w_i},{w_i}P) 加入到 {L_3} 中并输出 {w_i}P .

    4) {H_4} Hash询问.当 \mathcal{C} 收到 {\mathcal{A}_2} ({R_i},{H_4}({R_i})) 的查询后,若在 {L_4} 中已有 ({R_i},{H_4}({R_i})) 则返回 {H_4}({R_i}) {\mathcal{A}_2} ;否则, \mathcal{C} 选取 {H_4}({R_i}) \in {\{ 0,1\} ^{{l_0} + {l_1}}} ,并将 ({R_i},{H_4}({R_i})) 加入到 {L_4} 中且输出 {H_4}({R_i}) .

    5) {H_5} Hash询问.当 \mathcal{C} 收到 {\mathcal{A}_2} {f_{i,d}} 的查询,其中 d \in \{ 1,2, \cdot \cdot \cdot n\} ,若 {L_5} 存在 ({m_i},n,{f_{i,0}}, \cdot \cdot \cdot ,{f_{i,d - 2}},{f_{i,d}}) 则返回 {f_{i,d}} {\mathcal{A}_2} ;否则, \mathcal{C} 选取 {f_{i,*}} \in \mathbb{Z}_p^* ,将 ({m_i},n,{f_{i,0}}, \cdot \cdot \cdot ,{f_{i,d - 2}},{f_{i,d}}) 加入到 {L_5} 中并输出 {f_{i,d}} .

    6) {H_6} Hash询问.当 \mathcal{C} 收到 {\mathcal{A}_2} {C_{i,6}} 的查询后,若在 {L_6} 中已有 {C_{i,6}} 则返回 {C_{i,6}} {\mathcal{A}_2} ;否则, \mathcal{C} 选取 {C_{i,6}} \in {\{ 0,1\} ^k} ,将相应元组加入到 {L_6} 中并输出 {C_{i,6}} .

    7) 密钥提取询问.当 \mathcal{C} 收到 {\mathcal{A}_2} I{D_i} 的私钥的查询后, \mathcal{C} 首先查询 {L_1} 中是否存在 ({x_i},P{K_{i,1}},P{K_{i,2}},I{D_i}) ,若不存在则输出“ \bot ”;否则返回 ({x_i},P{K_{i,1}},*,*) .如果I{D_i} \notin \{ I{D_i}\} _{i = 1}^n \mathcal{C} I{D_i} 作为 {H_1} Hash询问的输入,得到 {Q_i} = {H_0} (I{D_i}) ,并计算 s{k_{i,1}} = a{Q_i} s{k_{i,2}} = {x_i} + aP{K_{i,2}} ,返回 (P{K_{i,1}}, s{k_{i,1}}, P{K_{i,2}},I{D_i}) {\mathcal{A}_2} .

    8) 公钥替换询问.当 \mathcal{C} 收到 {\mathcal{A}_2} (I{D_i},P{K_{i,1}},P{K_{i,2}}) 的查询后,若 ({x_i},P{K_{i,1}},P{K_{i,2}},I{D_i}) 已存在于 {L_1} 中,则 \mathcal{C} 用列表L1中的 (P{K_{i,1}},P{K_{i,2}}) 替换 I{D_i} 原有的公钥(P{K_{i,1}}, P{K_{i,2}});否则, \mathcal{C} ({x_i},P{K_{i,1}}, P{K_{i,2}},I{D_i}) 加入到列表 {L_1} 中.

    9) 签密询问.当 \mathcal{C} 收到 {\mathcal{A}_2} ({m_i},I{D_i},I{D_j}) 的询问后, \mathcal{C} 执行①~②操作:

    ① 若 I{D_i} \ne I{D_l} {\mathcal{A}_2} 没有对 I{D_i} 的公钥执行过替换询问, \mathcal{C} 通过 {H_1} Hash询问与密钥提取询问分别获取 {x_i} s{k_{i,2}} ,并对 {m_i} 进行签密;若 I{D_i} 对应的公钥被替换过, \mathcal{C} 首先通过 {H_1} 询问分别获取 (P{K_{i,1}},P{K_{i,2}}) (P{K_{j,1}},P{K_{j,2}}) ,然后 \mathcal{C} 利用随机数 {a_i} \in \mathbb{Z}_p^* 计算 {C_{i,1}} = {a_i}P {R_i} = {a_i}{Q_j}P_{{\text{pub}}}',并通过 {H_2} {H_3} {H_4} Hash询问分别获取 {U_i} = {H_2}({m_i}, I{D_i}, I{D_j}, {R_i},P{K_{i,1}},P{K_{j,1}}) {V_i} = {H_3}({m_i},I{D_i},I{D_j},{R_i},P{K_{i,1}},P{K_{j,1}}) . {H_4} ({R_i}) ,通过密钥提取询问获取私钥 s{k_{i,2}} ,计算 {v_i} = \ {a_i}{U_i} + s{k_{i,2}}{V_i} {C_{i,3}} = {v_i}P {C_{i,4}} = {H_4}({R_i}) \oplus ({m_i}||{v_i}) ,最后输出密文 {\sigma _i} = ({C_{i,1}},{C_{i,2}},{C_{i,3}},P{K_{i,1}}) {\mathcal{A}_2} .

    ② 若 I{D_i} = I{D_l} \mathcal{C} 首先通过 {H_1} 询问分别获取 (P{K_{i,1}}, P{K_{i,2}}) (P{K_{j,1}},P{K_{j,2}}) ,随机选择 y,z \in \mathbb{Z}_p^* 并计算 {C_{i,1}} = zaP .然后 \mathcal{C} 通过 {H_1} Hash询问和 {H_4} Hash询问分别获取 (I{D_j}, {a_j}) {H_4}({R_j}) ,并计算{R_j} = {a_j}{Q_j}P_{{\text{pub}}}' {U_j} = {H_2}({m_l},I{D_l},I{D_j}, {R_j}, P{K_{l,1}},P{K_{j,1}}) ,将 ({m_l},I{D_l},I{D_j},{R_j},P{K_{l,1}},P{K_{j,1}},{U_j}) 加入到 {L_2} 中,通过 {H_3} Hash询问获取 ({m_l},I{D_l},I{D_j},{R_l},P{K_{l,1}}, P{K_{j,1}}, {V_l},{w_l},{w_l}P) ,并计算 {v_l} = y{U_l} {C_{l,3}} = z{v_l}P_{{\text{pub}}}' + {w_l}P{K_{l,1}} {C_{i,4}} = {H_4} ({R_l}) \oplus ({m_l}||{v_l}) ,最后输出 {\sigma _l} = ({C_{l,1}},{C_{l,2}},{C_{l,3}},P{K_{l,1}}) {\mathcal{A}_2} .

    10) 解签密询问.当 \mathcal{C} 收到 {\mathcal{A}_2} (C{T_1},C{T_2}, \cdot \cdot \cdot , C{T_n}, \{ I{D_i}\} _{i = 1}^n,I{D_j}) 的查询后, \mathcal{C} 执行①~②操作:

    ① 对 (I{D_1},I{D_2}, \cdot \cdot \cdot ,I{D_n},I{D_j}) 分别执行 {H_1} Hash询问以获取 ({Q_1},{Q_2}, \cdot \cdot \cdot ,{Q_n},{Q_j}) (P{K_{1,1}},P{K_{2,1}}, \cdot \cdot \cdot ,P{K_{n,1}}, P{K_{j,1}}) ,然后 \mathcal{C} 执行聚合签名验证算法,若验证未通过,则输出“ \bot ”后终止模拟;否则继续执行后续操作.

    ② 若I{D_j} \ne I{D_l} \mathcal{C} 则通过 {H_1} Hash询问获取 (I{D_j}, {a_j}) 并计算 {R_j} = {a_j}{C_{j,1}} ,检查 {L_2} 中是否存在元组 (*,I{D_j},{R_i}, P{K_{i,1}},P{K_{j,1}},{U_i}) ,若存在,则 \mathcal{C} 利用Hash值 {U_i} 对密文进行解密;否则 \mathcal{C} 随机选取 {U_i} \in \mathbb{Z}_p^* 并用 {U_i} 对密文进行解密.若 I{D_j} = I{D_l} \mathcal{C} 则在 {L_2} 中查询是否存在元组(*,I{D_j},*, P{K_{i,1}},P{K_{j,1}},{U_i}),若存在则利用Hash值 {U_i} 对密文进行解密;否则将随机选取 {U_i} \in \mathbb{Z}_p^* 并用 {U_i} 对密文进行解密.

    11) 测试陷门询问.当 \mathcal{C} 收到 {\mathcal{A}_2} t{k_j} 的询问后,若 {L_1} 中存在元组 ({x_i},P{K_{i,1}},P{K_{i,2}},I{D_i}) \mathcal{C} 通过 {H_1} 询问获取s{k_{i,3}} ={H_1}(I{D_i}||s)并返回 t{k_j} = s{k_{i,3}} {\mathcal{A}_2} ;否则, \mathcal{C} 选取t{k_j} \in \mathbb{Z}_p^*发送给 {\mathcal{A}_2} ,并将 ({x_i},P{K_{i,1}},P{K_{i,2}},I{D_i}) 加入到 {L_{{\text{td}}}} 中.

    挑战阶段. {\mathcal{A}_2} 输出2个消息 m_0^* = \{ m_{i,0}^*\} _{i = 1}^n m_1^* = \{ m_{i,1}^*\} _{i = 1}^n ,并输出身份 \{ ID_i^*\} _{i = 1}^n ID_j^* \mathcal{C} ID_j^* 作为输入进行 {H_1} Hash询问,若 {L_1} 中不存在与 ID_j^* 相关的元组,则 \mathcal{C} 挑战失败;否则, \mathcal{C} {L_1} 中获取 \{ ID_i^*\} _{i = 1}^n 对应的公钥 \{ PK_{i,1}^*,PK_{i,2}^*\} _{i = 1}^n ,随机选择 \{ s{k_{i,2}} \in \mathbb{Z}_p^*\} _{i = 1}^n 并计算 \{ {C_{i,1}} = s{k_{i,2}}cP\} _{i = 1}^n ;然后 \mathcal{C} {L_2} {L_3} 中获取 \{ {U_i}\} _{i = 1}^n \{ {V_i}\} _{i = 1}^n ,并计算 v_i^* = {a_i}{U_i} + s{k_{i,2}}{V_i} = {t_i}C_{i,1}^* + s{k_{i,2}}{w_i}PK_{i,1}^* ,其中 {t_i} {w_i} s{k_{i,2}} 分别来自 {H_2} Hash询问、 {H_3} Hash询问与对 ID_j^* 的密钥提取询问;随后 \mathcal{C} 随机选择 \mu \in \{ 0,1\} 并计算 C_{i,4}^* = {H_4}({R_i}) \oplus ({m_{i,\mu }}||v_i^*) C_{i,3}^* = v_i^*P ,然后通过 {H_1} Hash询问获取公钥 \{ PK_{i,1}^*\} _{i = 1}^n 并输出 {\sigma ^*} = (C_{1,1}^*, \cdot \cdot \cdot ,C_{n,1}^*,C_{1,3}^*, \cdot \cdot \cdot ,C_{n,3}^*,C_{1,4}^*, \cdot \cdot \cdot ,C_{n,4}^*,PK_{1,1}^*, \cdot \cdot \cdot ,PK_{n,1}^*) {\mathcal{A}_2} .

    询问阶段2. {\mathcal{A}_2} 执行与询问阶段1类似的多项式有界次适应性查询,但不允许对 ID_i^* ID_j^* 对应的密文进行解签密查询.

    猜测阶段. {\mathcal{A}_2} 输出1个对 \mu 的猜测\mu {'} \in \{ 0,1\},如果\mu {'} = \mu,则 {\mathcal{A}_2} 在以上游戏中获胜. \mathcal{C} 在列表 {L_4} 中选取 ({R_i},{H_4}({R_i})) 并以 {R_i} = abP 作为CDH困难问题的解,这与目前公认的CDH问题的难解性相矛盾.因此本文方案在面对A2敌手时满足选择OW-CCA2安全. 证毕.

    将本文提出的方案与文献[2226]方案在功能特性方面进行比较,对比结果如表1所示.与文献[2324]方案相比,本文方案引入等值测试功能,实现了对存储在云端的医疗密文的安全检索.与文献[22,2526]方案相比,本文方案引入了聚合签密技术,确保了WBAN中医疗数据的机密性、完整性与可认证性,提高了多用户环境下对医疗数据进行签密与验证的效率.文献[2526]方案采用的等值测试方法只能对2个密文进行比较,本文方案实现了同时对多个密文进行匹配,降低了测试者执行密文等值测试时的开销.此外,与文献[2223,2526]方案相比,本文方案达到了适应性选择密文攻击下的单向性,安全性有所提升.

    表  1  功能特性比较
    Table  1.  Comparison of Functional Characteristics
    方案等值
    测试
    多密文等值
    测试
    签密聚合
    签密
    安全性
    文献[22]方案××选择明文攻击下的单向性
    文献[23]方案××选择密文攻击
    下的不可区分性
    文献[24]方案××适应性选择密文攻击
    下的不可区分性
    文献[25]方案×××选择密文攻击下的单向性
    文献[26]方案××选择密文攻击下的单向性
    本文方案适应性选择密文攻击
    下的单向性
    注:“×”表示不具有某种特定功能;“√”表示具有某种特定功能.
    下载: 导出CSV 
    | 显示表格

    本文所提新方案在执行多密文等值测试算法时,测试者通过对范德蒙矩阵求逆以提取出与数据拥有者明文相关的系数.其中,n阶范德蒙矩阵求逆算法的时间复杂度取决于所使用的求逆方法,已有许多学者提出了求解范德蒙矩阵逆矩阵的串行[27-28]与并行[29-30]方法,其时间复杂度如表2所示:

    表  2  范德蒙矩阵求逆算法复杂度
    Table  2.  Complexity of Inversion for Vandermonde Matrix
    方案时间复杂度
    文献[27]方案 O({n^2})
    文献[28]方案 O({n^2})
    文献[29]方案 O((\log n))
    文献[30]方案 O({(\log n)^2})
    下载: 导出CSV 
    | 显示表格

    将本文提出的方案在计算时间开销方面与文献[2526]方案进行对比,假设参与密文等值测试的用户数量为n,使用i7-8750h,2.20 GHz处理器,8 GB内存和Win10操作系统在VC6.0环境下用PBC库分别对本文方案与对比方案进行了仿真模拟,对比结果如表3所示.其中标量乘法运算时间Tsm = 0.0004 ms,群元素乘法运算时间Tmul = 0.0314 ms,Hash函数运算时间Th = 0.0001 ms,指数运算时间Te = 6.9866 ms,双线性配对时间Tbp = 9.6231 ms,范德蒙矩阵求逆时间Tinv取决于矩阵求逆方法.从表3可以看出,由于本文方案中不存在计算开销较大的双线性配对运算,因此在密文生成阶段的计算时间开销相比于文献[2526]的方案有显著降低.在数据解密及验证阶段,非聚合模式下的文献[2526]方案需要所有数据用户逐一对数据进行验证并解密,而本文方案中的数据用户能够对聚合密文进行批量验证,验证效率相比于文献[2526]的方案有所提高.

    表  3  计算量比较
    Table  3.  Computation Amount Comparison ms
    方案密文生成时间密文等值测试时间数据解密及验证时间
    文献[25]方案\begin{aligned} & n{T_{ {\text{mul} } } } + 3n{T_{ {\text{bp} } } } + 6n{T_{\text{h} } } + 5n{T_{\text{e} } } \\ &\quad( 63.8343n )\end{aligned}\begin{aligned} & (n - 1)(4{T_{ {\text{bp} } } } + 2{T_{\text{h} } }) \\ &\quad ( 38.4926n - 38.4926) \end{aligned}\begin{aligned} & 2n{T_{ {\text{bp} } } } + 4n{T_{\text{h} } } + 2n{T_{{\rm{e}} } }\\ &\quad (33.2198n) \end{aligned}
    文献[26]方案\begin{aligned} & 6n{T_{ {\text{sm} } } } + 2n{T_{ {\text{bp} } } } + 7n{T_{\text{h} } } + 2n{T_{\text{e} } } \\ &\quad( 33.2250n) \end{aligned}\begin{aligned} & (n - 1)(4{T_{ {\text{bp} } } } + 2{T_{\text{h} } }) \\ &\quad( 38.4926n - 38.4926) \end{aligned}\begin{aligned}& 3n{T_{ {\text{sm} } } } + n{T_{ {\text{mul} } } } + 5n{T_{ {\text{bp} } } } + 5n{T_{\text{h} } }\\ &\quad ( 48.1486n )\end{aligned}
    本文方案\begin{aligned} & 7n{T_{ {\text{sm} } } } + n{T_{ {\text{mul} } } } + n(n + 4){T_{\text{h} } }\\ &\quad ( 0.0346n + 0.0001{n^2})\end{aligned}\begin{aligned} & n{T_{ {\text{sm} } } } + 2n{T_{\text{h} } } + {T_{ {\text{inv} } } }\\ &\quad ( {T_{ {\text{inv} } } } + 0.0006n) \end{aligned}\begin{aligned} & n(2 + 4n){T_{ {\text{sm} } } } + {n^2}{T_{ {\text{mul} } } } + n(n + 4){T_{\text{h} } } \\ &\quad ( 0.0012n + 0.0331{n^2}) \end{aligned}
    注:n表示参与密文等值测试的用户数量;T_{\text{sm}}表示标量乘法运算时间;T_{\text{mul}}表示群元素乘法运算时间;T_{\text{h}}表示Hash函数运算时间;T_{\text{e}}表示指数运算时间;T_{\text{bp}}表示双线性配对时间;T_{\text{inv}}表示范德蒙矩阵求逆时间.
    下载: 导出CSV 
    | 显示表格

    此外,文献[2526]方案仅支持将多个用户的密文两两一组进行匹配,其密文等值测试算法中双线性配对运算数量与参与测试的用户数量呈线性关系;而本文方案中,测试者可以同时对 n 个用户的密文进行匹配,且测试过程中不存在双线性配对运算.本文方案的等值测试时间主要取决于测试者对范德蒙行列式求逆时所选取的算法,而在对范德蒙矩阵求逆的过程中仅进行标量加法与乘法等计算效率较高的运算[28],因此本文方案的密文等值测试效率同样高于文献[2526]方案的效率.

    针对现有的WBAN密码方案在多用户环境下计算效率较低等问题,本文提出了支持多密文等值测试的WBAN聚合签密方案.该方案采用基于身份的密码体制,消除了传统公钥方案中证书管理的开销;引入多密文等值测试技术,实现了多数据用户对多医疗密文的同时检索;减少了多用户环境下密文等值测试的计算开销;利用聚合签密技术,提高了对多个用户的医疗数据进行签密的效率.本文方案满足医疗数据在传输过程中的机密性、完整性和可认证性,同时保证了数据拥有者签名的不可伪造性与测试陷门的单向性.与同类方案的对比分析结果表明,本文方案支持更多安全属性且计算开销更低.在未来的工作中,将尝试设计抗量子计算攻击的支持多密文等值测试的WBAN签密方案.

    作者贡献声明:杨小东负责论文整体思路与实验方案的设计;周航负责设计方案与撰写论文;任宁宁负责方案仿真与效率分析;袁森负责搜集应用场景相关资料;王彩芬提出指导意见并修改论文.

  • 图  1   2018—2022年调研文献来源统计分析

    Figure  1.   Statistics analysis of research literature source from 2018 to 2022

    图  2   外包计算威胁模型

    Figure  2.   Threat model for outsourced computing

    图  3   安全外包计算通用模型

    Figure  3.   General model for secure outsourced computing

    图  4   按机器学习任务阶段分类的2种模式

    Figure  4.   Two modes classified by task phase of machine learning

    图  5   按云服务商数量分类的2种模式

    Figure  5.   Two modes classified by the number of cloud service providers

    图  6   机器学习安全外包研究模型统计

    Figure  6.   Model statistics of machine learning security outsourced research

    图  7   通过岭回归训练逻辑回归

    Figure  7.   Training logistic regression by ridge regression

    图  8   朴素贝叶斯分类模型托管框架

    Figure  8.   Naive Bayes classification model hosting framework

    图  9   用户身份认证机制

    Figure  9.   User authentication mechanism

    图  10   决策树转换

    Figure  10.   Decision tree transformation

    图  11   双云模式下的DNN推理

    Figure  11.   DNN inference in two cloud modes

    表  1   机器学习外包计算模型的特点

    Table  1   Features of Machine Learning Outsourced Computing Modes

    模式应用趋势特点
    由模型训练到模型托管
    支持多用户
    减少用户与云服务商交互
    支持密文托管
    支持用户离线
    支持模型机密性保护
    考虑半可信云服务器威胁
    由单云到多云分摊计算,支持MPC协议
    减少用户与云服务商交互
    支持用户离线
    增加恶意云服务器的威胁
    下载: 导出CSV

    表  2   逻辑回归外包方案对比

    Table  2   Comparison of Logistic Regression Outsourced Schemes

    来源单云/多云外包阶段加密工具Sigmoid函数逼近威胁模型支持SIMD
    文献[16]单云训练近似FHE最小二乘拟合多项式半可信
    文献[19]多云训练层次HE,SGX泰勒展开式半可信
    文献[20]单云托管FHE泰勒展开式半可信×
    文献[21]单云训练FHE最小二乘拟合多项式半可信
    文献[22]单云训练层次HE半可信
    下载: 导出CSV

    表  3   朴素贝叶斯分类外包方案对比

    Table  3   Comparison of Naive Bayesian Classification Outsourced Schemes

    来源单云/多云引入可信实体外包阶段加密工具威胁模型数据机密性模型机密性通信量
    文献[24]单云×推理Paillier半可信××
    文献[25]单云推理Paillier半可信
    文献[26]单云×推理OU恶意用户
    文献[28]单云×训练Paillier半可信
    文献[31]单云训练+推理Paillier半可信×
    下载: 导出CSV

    表  4   支持向量机外包方案对比

    Table  4   Comparison of Support Vector Machine Outsourced Schemes

    来源单云/多云加密模型引入可信实体外包阶段加密工具威胁模型结果可验证性多次交互
    文献[35]单云×推理Paillier+GC半可信×
    文献[36]多云训练+推理DT-PKC半可信××
    文献[37]多云训练DT-PKC半可信×
    文献[38]单云××推理FHE半可信××
    文献[40]多云推理矩阵盲化恶意云
    文献[41]单云×推理DT-PKC / BGN恶意用户×
    下载: 导出CSV

    表  5   决策树外包方案对比

    Table  5   Comparison of Decision Tree Outsourced Schemes

    来源单云/多云加密模型外包阶段加密工具威胁模型支持随机森林离线
    文献[46]单云×推理HE+ GC+OT半可信××
    文献[47]单云×推理GC, OT, ORAM半可信××
    文献[49]多云推理秘密共享半可信×
    文献[50]多云训练+推理FHE半可信
    文献[51]多云推理GC+秘密共享恶意用户×
    文献[52]单云推理对称加密半可信×
    文献[53]单云推理HE恶意用户××
    文献[54]多云推理HE+秘密共享半可信×
    文献[55]多云×训练+推理DT-PKC+秘密共享半可信×
    文献[56]单云推理多密钥HE+OT半可信×
    文献[57]单云训练+推理矩阵盲化半可信×
    文献[58]单云推理HE半可信×
    下载: 导出CSV

    表  6   神经网络外包方案对比

    Table  6   Comparison of Neural Network Outsourced Schemes

    来源算法单云/多云外包阶段加密工具威胁模型可验证模型隐私
    文献[60]SLP单云训练+推理矩阵盲化半可信×
    文献[61]SLP单云训练+推理矩阵盲化恶意云
    文献[62]DNN单云训练+推理DT-PKC半可信×
    文献[63]R-CNN多云训练秘密共享半可信×
    文献[64]DNN单云训练加噪半可信××
    文献[65]CNN多云训练秘密共享半可信××
    文献[66]DNN单云训练矩阵盲化半可信×
    文献[67]DNN多云推理秘密共享半可信×
    文献[68]DNN多云推理秘密共享半可信××
    文献[69]CNN多云推理秘密共享半可信××
    文献[70]DNN多云推理秘密共享半可信×
    文献[71]CNN单云推理加性HE半可信××
    文献[72]DNN单云推理HE+ GC+秘密共享半可信×
    文献[73]CNN单云推理HE+秘密共享半可信×
    文献[74]CNN单云推理HE+ GC+秘密共享恶意用户×
    文献[75]CNN多云推理GC+秘密共享半可信×
    文献[76]DNN多云推理GC恶意云×
    文献[77]DNN多云推理HE+秘密共享半可信×
    文献[78]DNN多云训练+推理GC+秘密共享半可信×
    文献[7980]DNN多云推理秘密共享半可信×
    文献[8184]CNN多云推理GC+秘密共享恶意云×
    下载: 导出CSV
  • [1] 李印,陈勇,赵景欣,等. 泛在计算安全综述[J]. 计算机研究与发展,2022,59(5):1054−1081 doi: 10.7544/issn1000-1239.20211248

    Li Yin, Chen Yong, Zhao Jingxin, et al. Survey of ubiquitous computing security[J]. Journal of Computer Research and Development, 2022, 59(5): 1054−1081 (in Chinese) doi: 10.7544/issn1000-1239.20211248

    [2] 周俊,沈华杰,林中允,等. 边缘计算隐私保护研究进展[J]. 计算机研究与发展,2020,57(10):2027−2051 doi: 10.7544/issn1000-1239.2020.20200614

    Zhou Jun, Shen Huajie, Lin Zhongyun, et al. Research advances on privacy preserving in edge computing[J]. Journal of Computer Research and Development, 2020, 57(10): 2027−2051 (in Chinese) doi: 10.7544/issn1000-1239.2020.20200614

    [3]

    Statista. Public cloud services end-user spending worldwide from 2017 to 2023[EB/OL]. [2022-08-15]. https://www.statista.com/statistics/273818/global-revenue-generated-with-cloud-computing-since-2009/

    [4]

    Shan Zihao, Ren Kui, Blanton M, et al. Practical secure computation outsourcing: A survey[J]. ACM Computing Surveys, 2019, 51(2): 1−40

    [5]

    Zhou Lei, Fu Anmin, Yang Guomin, et al. Efficient certificateless multi-copy integrity auditing scheme supporting data dynamics[J]. IEEE Transactions on Dependable and Secure Computing, 2022, 19(2): 1118−1132

    [6]

    Liu Bo, Ding Ming, Shaham S, et al. When machine learning meets privacy: A survey and outlook[J]. ACM Computing Surveys, 2022, 54(2): 1−36

    [7] 李帅,付安民,苏铓,等. 基于单服务器的群上幂指数安全外包计算方案[J]. 计算机研究与发展,2018,55(11):2482−2489 doi: 10.7544/issn1000-1239.2018.20170420

    Li Shuai, Fu Anmin, Su Mang, et al. Secure and verifiable protocol for outsourcing group power exponent to a single server[J]. Journal of Computer Research and Development, 2018, 55(11): 2482−2489 (in Chinese) doi: 10.7544/issn1000-1239.2018.20170420

    [8] 聂恒太,王少辉. 云环境下矩阵乘法外包计算方案[J]. 计算机技术与发展,2018,28(8):119−123 doi: 10.3969/j.issn.1673-629X.2018.08.025

    Nie Hengtai, Wang Shaohui. A matrix multiplication outsourcing calculation scheme in cloud environment[J]. Computer Technology and Development, 2018, 28(8): 119−123 (in Chinese) doi: 10.3969/j.issn.1673-629X.2018.08.025

    [9]

    Fu Anmin, Chen Zhenzhu, Mu Yi, et al. Cloud-based outsourcing for enabling privacy-preserving large-scale non-negative matrix factorization[J]. IEEE Transactions on Services Computing, 2022, 15(1): 266−278 doi: 10.1109/TSC.2019.2937484

    [10] 李晓伟,陈本辉,杨邓奇,等. 边缘计算环境下安全协议综述[J]. 计算机研究与发展,2022,59(4):765−780 doi: 10.7544/issn1000-1239.20210644

    Li Xiaowei, Chen Benhui, Yang Dengqi, et al. Review of security protocols in edge computing environments[J]. Journal of Computer Research and Development, 2022, 59(4): 765−780 (in Chinese) doi: 10.7544/issn1000-1239.20210644

    [11]

    Kumar M, Sharma S C, Goel A, et al. A comprehensive survey for scheduling techniques in cloud computing[J]. Journal of Network and Computer Applications, 2019, 143: 1−33 doi: 10.1016/j.jnca.2019.06.006

    [12]

    Domingo-Ferrer J, Farras O, Ribes-González J, et al. Privacy-preserving cloud computing on sensitive data: A survey of methods, products and challenges[J]. Computer Communications, 2019, 140: 38−60

    [13]

    Shama T, Wang Tian, Giulio C D, et al. Towards inclusive privacy protections in the cloud[C]//Prof of the 18th Int Conf on Applied Cryptography and Network Security. Berlin: Springer, 2020: 337−359

    [14] 周纯毅,陈大卫,王尚,等. 分布式深度学习隐私与安全攻击研究进展与挑战[J]. 计算机研究与发展,2021,58(5):927−943 doi: 10.7544/issn1000-1239.2021.20200966

    Zhou Chunyi, Chen Dawei, Wang Shang, et al. Research and challenge of distributed deep learning privacy and security attack[J]. Journal of Computer Research and Development, 2021, 58(5): 927−943 (in Chinese) doi: 10.7544/issn1000-1239.2021.20200966

    [15]

    Kleinbaum D G, Klein M. Logistic Regression: A Self-learning Text[M]. Berlin: Springer, 2010

    [16]

    Han K, Hong S, Cheon J H, et al. Logistic regression on homomorphic encrypted data at scale[C]//Proc of the 31st AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2019: 9466−9471

    [17]

    Bos J W, Lauter K, Naehrig M. Private predictive analysis on encrypted medical data[J]. Journal of Biomedical Informatics, 2014, 50: 234−243 doi: 10.1016/j.jbi.2014.04.003

    [18]

    Aono Y, Hayashi T, Trieu P L, et al. Scalable and secure logistic regression via homomorphic encryption[C]//Proc of the 6th ACM Conf on Data and Application Security and Privacy. New York: ACM, 2016: 142−144

    [19]

    Jiang Yichen, Hamer J, Wang Chenghong, et al. SecureLR: Secure logistic regression model via a hybrid cryptographic protocol[J]. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2018, 16(1): 113−123

    [20]

    Fan Yongkai, Bai Jianrong, Lei Xia, et al. Privacy preserving based logistic regression on big data[J]. Journal of Network and Computer Applications, 2020, 171: 1−10

    [21]

    Yu Xiaopeng, Zhao Wei, Huang Yunfan, et al. Privacy-preserving outsourced logistic regression on encrypted data from homomorphic encryption[J]. Security and Communication Networks, 2022, 2022: 1−17

    [22]

    Byun J, Lee W, Lee J. Parameter-free HE-friendly logistic regression[J]. Advances in Neural Information Processing Systems, 2021, 34: 8457−8468

    [23]

    Murphy K P. Naive Bayes classifiers[EB/OL]. [2022-08-15]. https://www.cs.ubc.ca/~murphyk/Teaching/CS340-Fall06/reading/NB.pdf

    [24]

    Li Tong, Huang Zhengan, Li Ping, et al. Outsourced privacy-preserving classification service over encrypted data[J]. Journal of Network and Computer Applications, 2018, 106: 100−110 doi: 10.1016/j.jnca.2017.12.021

    [25]

    Li Tong, Li Xuan, Zhong Xingyi, et al. Communication-efficient outsourced privacy-preserving classification service using trusted processor[J]. Information Sciences, 2019, 505: 473−486 doi: 10.1016/j.ins.2019.07.047

    [26]

    Chai Yanting, Zhan Yu, Wang Baocang, et al. Improvement on a privacy-preserving outsourced classification protocol over encrypted data[J]. Wireless Networks, 2020, 26(6): 4363−4374 doi: 10.1007/s11276-020-02329-9

    [27]

    Okamoto T, Uchiyama S. A new public-key cryptosystem as secure as factoring[C]//Proc of the 17th Int Conf on the Theory and Applications of Cryptographic Techniques. Berlin: Springer, 1998: 308−318

    [28]

    Li Tong, Li Jin, Liu Zheli, et al. Differentially private naive Bayes learning over multiple data sources[J]. Information Sciences, 2018, 444: 89−104 doi: 10.1016/j.ins.2018.02.056

    [29]

    Vaidya J, Shafiq B, Basu A, et al. Differentially private naive Bayes classification[C]//Proc of 2013 IEEE/WIC/ACM Int Joint Conf on Web Intelligence and Intelligent Agent Technologies. Piscataway, NJ: IEEE, 2013: 571−576

    [30]

    Huai Mengdi, Huang Liusheng, Yang Wei, et al. Privacy-preserving naive Bayes classification[C]//Proc of the 8th Int Conf on Knowledge Science, Engineering and Management. Berlin: Springer, 2015: 627−638

    [31]

    Wang Fengwei, Zhu Hui, Lu Rongxing, et al. Achieve efficient and privacy-preserving disease risk assessment over multi-outsourced vertical datasets[J]. IEEE Transactions on Dependable and Secure Computing, 2020, 19(33): 1492−1504

    [32]

    Pisner D A, Schnyer D M. Support Vector Machine[M]. New York: Academic Press, 2020

    [33]

    Lin K P, Chen M S. Privacy-preserving outsourcing support vector machines with random transformation[C]//Proc of the 16th ACM SIGKDD Int Conf on Knowledge Discovery and Data Mining. New York: ACM, 2010: 363−372

    [34]

    Rahulamathavan Y, Phan R C W, Veluru S, et al. Privacy-preserving multi-class support vector machine for outsourcing the data classification in cloud[J]. IEEE Transactions on Dependable and Secure Computing, 2014, 11(5): 467−479 doi: 10.1109/TDSC.2013.51

    [35]

    Li Xingxin, Zhu Youwen, Wang Jian, et al. On the soundness and security of privacy-preserving SVM for outsourcing data classification[J]. IEEE Transactions on Dependable and Secure Computing, 2017, 15(5): 906−912

    [36]

    Liu Ximeng, Deng R H, Choo K K R, et al. Privacy-preserving outsourced support vector machine design for secure drug discovery[J]. IEEE Transactions on Cloud Computing, 2018, 8(2): 610−622

    [37]

    Wang Jing, Wu Libing, Wang Huaqun, et al. An efficient and privacy-preserving outsourced support vector machine training for Internet of medical things[J]. IEEE Internet of Things Journal, 2020, 8(1): 458−473

    [38]

    Huang Hai, Wang Yongjian, Zong Haoren. Support vector machine classification over encrypted data[J]. Applied Intelligence, 2022, 52(6): 5938−5948 doi: 10.1007/s10489-021-02727-2

    [39]

    Barnett A, Santokhi J, Simpson M, et al. Image classification using non-linear support vector machines on encrypted data[EB/OL]. [2022-08-15].https://eprint.iacr.org/2017/857.pdf

    [40]

    Shao Yuhang, Tian Chengliang, Han Lidong, et al. Privacy-preserving and verifiable cloud-aided disease diagnosis and prediction with hyperplane decision-based classifier[J]. IEEE Internet of Things Journal, 2022, 9(21): 21648−21661 doi: 10.1109/JIOT.2022.3181734

    [41]

    Chen Yange, Mao Qinyu, Wang Baocang, et al. Privacy-preserving multi-class support vector machine model on medical diagnosis[J]. IEEE Journal of Biomedical and Health Informatics, 2022, 26(7): 3342−3353 doi: 10.1109/JBHI.2022.3157592

    [42]

    Charbuty B, Abdulazeez A. Classification based on decision tree algorithm for machine learning[J]. Journal of Applied Science and Technology Trends, 2021, 2(1): 20−28 doi: 10.38094/jastt20165

    [43]

    Bost R, Popa R A, Tu S, et al. Machine learning classification over encrypted data[C/OL]//Proc of the 22nd Symp on Network and Distributed System Security. Piscataway, NJ: IEEE, 2015[2022-08-15]. https://www.ndss-symposium.org/wp-content/uploads/2017/09/04_1_2.pdf

    [44]

    Wu D J, Feng T, Naehrig M, et al. Privately evaluating decision trees and random forests[C]//Proc of Privacy Enhancing Technologies Symp. Berlin: Springer, 2016: 335−355

    [45]

    Tai R K H, Ma J P K, Zhao Yongjun, et al. Privacy-preserving decision trees evaluation via linear functions[C]//Proc of European Symp on Research in Computer Security. Berlin: Springer, 2017: 494−512

    [46]

    Kiss Á, Naderpour M, Liu Jian, et al. Sok: Modular and efficient private decision tree evaluation[C]//Proc of Privacy Enhancing Technologies Symp. Berlin: Springer, 2019: 187−208

    [47]

    Tueno A, Kerschbaum F, Katzenbeisser S. Private evaluation of decision trees using sublinear cost[C]//Proc of Privacy Enhancing Technologies Symp. Berlin: Springer, 2019: 266−286

    [48]

    Zheng Yifeng, Duan Huayi, Wang Cong. Towards secure and efficient outsourcing of machine learning classification[C]//Proc of the 24th European Symp on Research in Computer Security. Berlin: Springer, 2019: 22−40

    [49]

    Zheng Yifeng, Duan Huayi, Wang Cong, et al. Securely and efficiently outsourcing decision tree inference[J]. IEEE Transactions on Dependable and Secure Computing, 2022, 19(3): 1841−1855 doi: 10.1109/TDSC.2020.3040012

    [50]

    Akavia A, Leibovich M, Resheff Y S, et al. Privacy-preserving decision trees training and prediction[J]. ACM Transactions on Privacy and Security, 2022, 25(3): 1−30

    [51]

    Ma J P K, Zhao Yongjun, Tai R K H. Let's stride blindfolded in a forest: Sublinear multi-client decision trees evaluation[C/OL]//Proc of the 28th Symp on Network and Distributed System Security. Piscataway, NJ: IEEE, 2021[2022-08-15]. https://www.ndss-symposium.org/wp-content/uploads/ndss2021_5C-1_23166_paper.pdf

    [52]

    Liang Jinwen, Qin Zheng, Xiao Sheng, et al. Efficient and secure decision tree classification for cloud-assisted online diagnosis services[J]. IEEE Transactions on Dependable and Secure Computing, 2021, 18(4): 1632−1644 doi: 10.1109/TDSC.2019.2922958

    [53]

    Wang Chen, Wang Andi, Xu Jian, et al. Outsourced privacy-preserving decision tree classification service over encrypted data[J]. Journal of Information Security and Applications, 2020, 53: 1−13

    [54]

    Liu Liu, Su Jinshu, Zhao Baokang, et al. Towards an efficient privacy-preserving decision tree evaluation service in the Internet of things[J]. Symmetry, 2020, 12(1): 1−16

    [55]

    Liu Liu, Chen Rongmao, Liu Ximeng, et al. Towards practical privacy-preserving decision tree training and evaluation in the cloud[J]. IEEE Transactions on Information Forensics and Security, 2020, 15: 2914−2929 doi: 10.1109/TIFS.2020.2980192

    [56]

    Aloufi A, Hu Peizhao, Wong H W H, et al. Blindfolded evaluation of random forests with multi-key homomorphic encryption[J]. IEEE Transactions on Dependable and Secure Computing, 2021, 18(4): 1821−1835

    [57]

    Wang Qinfan, Cui Shujie, Zhou Lei, et al. EnclaveTree: Privacy-preserving data stream training and inference using TEE[C]//Proc of the 17th ACM Asia Conf on Computer and Communications Security. New York: ACM, 2022: 741−755

    [58]

    Bai Jianli, Song Xiangfu, Cui Shujie, et al. Scalable private decision tree evaluation with sublinear communication[C]//Proc of the 17th ACM Asia Conf on Computer and Communications Security. New York: ACM, 2022: 843−857

    [59]

    Jain A K, Mao Jianchang, Mohiuddin K M. Artificial neural networks: A tutorial[J]. Computer, 1996, 29(3): 31−44 doi: 10.1109/2.485891

    [60]

    Zhang Chuan, Zhu Liehuang, Xu Chang, et al. PPDP: An efficient and privacy-preserving disease prediction scheme in cloud-based e-Healthcare system[J]. Future Generation Computer Systems, 2018, 79: 16−25 doi: 10.1016/j.future.2017.09.002

    [61]

    Zhang Xiaoyu, Chen Xiaofeng, Wang Jianfeng, et al. Verifiable privacy-preserving single-layer perceptron training scheme in cloud computing[J]. Soft Computing, 2018, 22(23): 7719−7732 doi: 10.1007/s00500-018-3233-7

    [62]

    Ma Xindi, Ma Jianfeng, Li Hui, et al. PDLM: Privacy-preserving deep learning model on cloud with multiple keys[J]. IEEE Transactions on Services Computing, 2018, 14(4): 1251−1263

    [63]

    Liu Yang, Ma Zhuo, Liu Ximeng, et al. Privacy-preserving object detection for medical images with faster R-CNN[J]. IEEE Transactions on Information Forensics and Security, 2019, 17: 69−84

    [64]

    Osia S A, Shamsabadi A S, Sajadmanesh S, et al. A hybrid deep learning architecture for privacy-preserving mobile analytics[J]. IEEE Internet of Things Journal, 2020, 7(5): 4505−4518 doi: 10.1109/JIOT.2020.2967734

    [65]

    Wagh S, Gupta D, Chandran N. SecureNN: 3-Party secure computation for neural network training[C]//Proc of Privacy Enhancing Technologies Symp. Berlin: Springer, 2019: 26−49

    [66]

    Melissourgos D, Gao Hanzhi, Ma Chaoyi, et al. On outsourcing artificial neural network learning of privacy-sensitive medical data to the cloud[C]//Prof of the 33rd Int Conf on Tools with Artificial Intelligence. Piscataway, NJ: IEEE, 2021: 381−385

    [67]

    Shamsabadi A S, Gascón A, Haddadi H, et al. PrivEdge: From local to distributed private training and prediction[J]. IEEE Transactions on Information Forensics and Security, 2020, 15: 3819−3831

    [68]

    Liu Xiaoning, Zheng Yifeng, Yuan Xingliang, et al. MediSC: Towards secure and lightweight deep learning as a medical diagnostic service[C]//Prof of the 26th European Symp on Research in Computer Security. Berlin: Springer, 2021: 519−541

    [69]

    Huang Kai, Liu Ximeng, Fu Shaojing, et al. A lightweight privacy-preserving CNN feature extraction framework for mobile sensing[J]. IEEE Transactions on Dependable and Secure Computing, 2021, 18(3): 1441−1455

    [70]

    Zheng Yifeng, Duan Huayi, Tang Xiaoting, et al. Denoising in the dark: Privacy-preserving deep neural network-based image denoising[J]. IEEE Transactions on Dependable and Secure Computing, 2021, 18(3): 1261−1275 doi: 10.1109/TDSC.2019.2907081

    [71]

    Juvekar C, Vaikuntanathan V, Chandrakasan A. GAZELLE: A low latency framework for secure neural network inference[C]//Proc of the 28th USENIX Security Symp. Berkeley, CA: USENIX Association, 2018: 1651−1669

    [72]

    Mishra P, Lehmkuhl R, Srinivasan A, et al. Delphi: A cryptographic inference service for neural networks[C]//Proc of the 29th USENIX Security Symp. Berkeley, CA: USENIX Association, 2020: 2505−2522

    [73]

    Zhang Qiao, Xin Chunsheng, Wu Hongyi. GALA: Greedy computation for linear algebra in privacy-preserved neural networks[C/OL]//Proc of the 28th Symp on Network and Distributed System Security. Piscataway, NJ: IEEE, 2021 [2022-08-15]. https://www.ndss-symposium.org/wp-content/uploads/ndss2021_5C-3_24351_paper.pdf

    [74]

    Lehmkuhl R, Mishra P, Srinivasan A, et al. Muse: Secure inference resilient to malicious clients[C]//Proc of the 30th USENIX Security Symp. Berkeley, CA: USENIX Association, 2021: 2201−2218

    [75]

    Riazi M S, Weinert C, Tkachenko O, et al. Chameleon: A hybrid secure computation framework for machine learning applications[C]//Proc of the 13th ACM Asia Conf on Computer and Communications Security. New York: ACM, 2018: 707−721

    [76]

    Riazi M S, Samragh M, Chen Hao, et al. XNOR: XNOR-based oblivious deep neural network inference[C]//Proc of the 28th USENIX Security Symp. Berkeley, CA: USENIX Association, 2019: 1501−1518

    [77]

    Rathee D, Rathee M, Kumar N, et al. CrypTFlow2: Practical 2-party secure inference[C]//Proc of ACM SIGSAC Conf on Computer and Communications Security. New York: ACM, 2020: 325−342

    [78]

    Agrawal N, Shahin S A, Kusner M J, et al. QUOTIENT: Two-party secure neural network training and prediction[C]//Proc of ACM SIGSAC Conf on Computer and Communications Security. New York: ACM, 2019: 1231−1247

    [79]

    Liu Xiaoning, Wu Bang, Yuan Xingliang, et al. Leia: A lightweight cryptographic neural network inference system at the edge[J]. IEEE Transactions on Information Forensics and Security, 2021, 17: 237−252

    [80]

    Liu Xiaoning, Zheng Yifeng, Yuan Xingliang, et al. Securely outsourcing neural network Inference to the cloud with lightweight techniques[J/OL]. IEEE Transactions on Dependable and Secure Computing, 2022[2022-10-25]. https://ieeexplore.ieee.org/document/9674792

    [81]

    Chaudhari H, Rachuri R, Suresh A. Trident: Efficient 4PC framework for privacy preserving machine learning[C/OL]//Proc of the 27th Symp on Network and Distributed System Security. Piscataway, NJ: IEEE, 2020[2022-08-15]. https://www.ndss-symposium.org/wp-content/uploads/2020/02/23005-paper.pdf

    [82]

    Mohassel P, Rindal P. ABY3: A mixed protocol framework for machine learning[C]//Proc of ACM SIGSAC Conf on Computer and Communications Security. New York: ACM, 2018: 35−52

    [83]

    Koti N, Pancholi M, Patra A, et al. SWIFT: Super-fast and robust privacy-preserving machine learning[C]//Proc of the 30th USENIX Security Symp. Berkeley, CA: USENIX Association, 2021: 2651−2668

    [84]

    Koti N, Patra A, Rachuri R, et al. Tetrad: Actively secure 4PC for secure training and inference[C/OL]//Proc of the 29th Symp on Network and Distributed System Security. Piscataway, NJ: IEEE, 2022[2022-08-15]. https://www.ndss-symposium.org/wp-content/uploads/2022−120-paper.pdf

    [85]

    Liu Yingqi, Ma Shiqing, Aafer Y, et al. Trojaning attack on neural networks[C/OL]//Proc of the 25th Symp on Network and Distributed System Security. Piscataway, NJ: IEEE, 2018[2022-08-15]. https://www.ndss-symposium.org/wp-content/uploads/2018/02/ndss2018_03A-5_Liu_paper.pdf

  • 期刊类型引用(0)

    其他类型引用(2)

图(11)  /  表(6)
计量
  • 文章访问数:  405
  • HTML全文浏览量:  95
  • PDF下载量:  203
  • 被引次数: 2
出版历程
  • 收稿日期:  2022-08-28
  • 修回日期:  2023-01-17
  • 网络出版日期:  2023-04-17
  • 刊出日期:  2023-06-30

目录

/

返回文章
返回