一种适合云数据共享的身份代理重加密方案

目前云数据安全存储方案中,数据拥有者加密数据上传到云中,但却不能很好的支持加密数据分享,尤其是分享给多个用户时,可扩展性不强。针对这个问题本文提出一种基于身份的代理重加密方案,该方案不需要云完全可信但却又能灵活地进行数据安全共享。在具体构造上,结合基于身份的加密,用一个强不可伪造的一次签名方案使被转化后的密文具有公开验证性,且能达到被转化后的密文在标准模型下具有选择密文安全性。由于该类方案无需使用公钥证书、能支持细粒度的访问控制且可扩展性较好,因此可以较好的适用于安全云数据共享。     

一种基于重复数据删除的镜像文件存储方法研究

在云计算环境中,基础设施即服务的日益发展导致虚拟机和虚拟机镜像的急剧增加,例如,Amazon Elastic Compute Cloud(EC2)有6521个公共虚拟机镜像文件,这给云环境的管理带来了极大的挑战,特别是大量镜像文件带来的重复数据的空间存储问题。为解决这一问题,提出一种基于固定分块的镜像文件重复数据删除的存储方案。当存储一个镜像文件时,先计算该镜像文件的指纹,并与指纹库的指纹比较,若存在则用指针替代,否则采用固定分块对镜像文件分割存储。为此,可以设计镜像文件元数据格式和镜像文件MD5索引表来解决上述问题。实验结果表明,内容相同的镜像文件只是元数据的开销并实现秒传,而相同版本、相同系统、不同软件的镜像组的重删率约达到58%。因此,本方案是非常有效的。     

EFADS: Efficient,flexible and anonymous data sharing protocol for cloud computing with proxy re-encryption

The concept of cloud computing has emerged as the next generation of computing infrastructure to reduce the costs associated with the management of hardware and software resources. It is vital to its success that cloud computing is featured efficient, flexible and secure characteristics. In this paper, we propose an efficient and anonymous data sharing protocol with flexible sharing style, named EFADS, for outsourcing data onto the cloud. Through formal security analysis, we demonstrate that EFADS provides data confidentiality and data sharer's anonymity without requiring any fully-trusted party. From experimental results, we show that EFADS is more efficient than existing competing approaches. Furthermore, the proxy re-encryption scheme we propose in this paper may be independent of interests, i.e., compared to those previously reported proxy re-encryption schemes, the proposed scheme is the first pairing-free, anonymous and unidirectional proxy re-encryption scheme in the standard model.     

云存储中支持数据去重的群组数据持有性证明

数据持有性证明(Provable Data Possession,简称PDP)和数据可恢复性证明 (Proofs of Retrievability,简称POR)是客户端用来验证存储在云端服务器上的数据完整性的主要技术.近几年它在学术界和工业界的应用广泛,很多PDP和POR方案相继出现,但是由于不同群组的特殊性和独特要求,使得群组PDP/POR方案多样化,并且群组应用中的许多重要功能(例如数据去重)没有被实现,如何构造高效及满足群组特定功能和安全需求的PDP/POR方案已经引起了人们的广泛关注.本文给出了一个支持数据去重的群组PDP方案(GPDP),基于矩阵计算和伪随机函数,GPDP可以在支持数据去重的基础上,高效的完成数据持有性证明,并且可以在群组中抵抗恶意方选择成员攻击.在标准模型下证明了GPDP的安全性,并且在百度云平台上实现了GPDP的原型系统.为了评估方案的性能,我们使用了10GB的数据量进行实验和分析,结果表明GPDP方案在达到群组中数据去重的目标基础上,可以高效地保证抵抗选择攻击和数据持有性,即预处理效率高于私有验证方案,而验证效率高于公开验证方案(与私有验证效率几乎相同).另外,与其他群组PDP/POR方案相比,GPDP方案将额外存储代价和通信代价都降到了最低.     

基于MapReduce的云存储数据审计方法研究

云存储是一种新兴的网络存储技术,它是云计算提供的一个重要服务。云存储因其快速、廉价和方便而广受云用户喜爱。然而,它也给云用户的外包数据带来了许多安全问题。其中一个重要问题就是如何确保半可信云服务器上数据的完整性。因此,云用户和云服务器亟需一个稳定、安全、可信的数据审计方法。随着大数据时代的到来,传统数据审计方案批量处理云环境下海量数据的效率不高;并且,随着移动客户端的流行,传统数据审计方案带给用户的在线负担太过繁重。因此,提出一种基于MapReduce编程框架的云数据审计方案,使用代理签名技术将用户对数据签名计算代理出去,并且并行化处理数据签名和批量审计过程。实验结果表明,所提方法明显提高了批量审计的效率,增强了云存储服务的可用性,并且减轻了用户的在线负担。     

一种支持完美隐私保护的批处理数据拥有性证明方案

数据拥有性证明技术是当前云存储安全领域中的一项重要研究内容,可使用户无须下载所有文件就能高效地远程校验用户数据是否完整存储于云服务器。现实中,用户趋向于委托第三方验证机构TPA代替自己来验证数据的完整性;然而,多数支持第三方公开审计的数据拥有性证明方案通常只考虑恶意服务器是否能够伪造标签或证明的问题,鲜有考虑恶意TPA可能会窃取用户隐私的情况。近几年,一些既针对服务器保证数据的安全性又针对TPA实现数据隐私保护的数据拥有性证明方案逐渐被提出,但多应用于单云服务器环境下;个别应用在多云服务器环境下可支持批量审计的方案,或者不能有效抵抗恶意云服务器的攻击,或者无法实现针对TPA的零知识隐私保护。因此,文中在Yu等工作的基础上,提出了一个多云服务器环境下支持批量审计的数据拥有性证明方案。所提方案既可保证针对恶意云服务器的安全性,还可实现针对TPA的完美零知识隐私保护。性能分析及仿真实验表明所提方案是高效且可行的。     

Chosen-ciphertext secure bidirectional proxy re-encryption schemes without pairings

Proxy re-encryption (PRE) is a useful primitive that allows a semi-trusted proxy to transform a ciphertext encrypted under one key into an encryption of the same plaintext under another key. A PRE scheme is bidirectional if the proxy is able to transform ciphertexts in both directions. In ACM CCS’07, Canetti and Hohenberger presented a bidirectional PRE scheme with chosen-ciphertext security, which captures the indistinguishability of ciphertexts even if the adversary has access to the standard decryption oracle as well as a re-encryption oracle and a re-encryption key oracle. They also left an important open problem to come up with a chosen-ciphertext secure PRE scheme without pairings. To resolve this problem, we propose a bidirectional PRE scheme without pairings, and prove its chosen-ciphertext security under the computational Diffie-Hellman assumption in the random oracle model. Based on this scheme, we further present a non-transitive bidirectional PRE scheme, in which the proxy alone cannot re-delegate the decryption rights.     

Efficient bidirectional proxy re-encryption with direct chosen-ciphertext security

Proxy re-encryption (PRE) allows a semi-trusted proxy to convert a ciphertext originally intended for a user into another ciphertext of the same message intended for another user, and the proxy, however, cannot learn anything about the message encrypted. In previous papers, in order to achieve the CCA2-security, a common method for constructing PRE schemes was to apply the paradigm of using strongly-unforgeable one-time signature which transforms a selective-identity, CPA-secure identity-based encryption (IBE) scheme into a CCA-secure cryptosystem. In this paper, we propose a direct design of the bidirectional CCA-secure PRE scheme, which makes a direct use of the underlying IBE structure and does not need any auxiliary signature mechanism. Our construction is efficient and suitable for further designing multi-user PRE schemes. Its security is proved on the base of the decisional bilinear Diffie-Hellman assumption in the standard model.     

Proxy re-encryption with keyword search

We introduce a new cryptographic primitive, called proxy re-encryption with keyword search, which is motivated by the following scenario in email systems: Charlie sends an encrypted email, which contains some keywords, such as “urgent”, to Alice under Alice’s public key, and Alice delegates her decryption rights to Bob via her mail server. The desired situations are: (1) Bob can decrypt mails delegated from Alice by using only his private key, (2) Bob’s mail gateway, with a trapdoor from Bob, can test whether the email delegated from Alice contains some keywords, such as “urgent”, (3) Alice and Bob do not wish to give the mail server or mail gateway the access to the content of emails.The function of proxy re-encryption with keyword search (PRES) is the combination of proxy re-encryption (PRE) and public key encryption with keyword search (PEKS). However, a PRES scheme cannot be obtained by directly combining those two schemes, since the resulting scheme is no longer proven secure in our security model. In this paper, a concrete construction is proposed, which is proven secure in the random oracle model, based on the modified Decisional Bilinear Diffie-Hellman assumption.     

云存储中基于虚拟用户的数据完整性验证

针对验证数据完整性过程中被撤销用户与云服务器存在共谋的问题,提出基于虚拟用户的数据完整性校验方案。在管理群组用户的过程中,管理员让云服务器作为代理,通过重签名方法将被撤销用户的签名转换为虚拟用户签名,以防止攻击者获取群组用户身份隐私信息。另一方面,管理员在本地存储所有用户的身份隐私信息,用户在访问共享数据之前需要通过管理员的验证,这样既能保证校验者可以正确验证共享数据的完整性,又能保护群组用户的隐私和共享数据的安全。分析证明结果表明,所提方案在用户撤销时不仅能够验证共享数据的完整性,还能降低攻击者精确获取用户身份隐私信息和共享数据内容的概率。     

Multi-use and unidirectional identity-based proxy re-encryption schemes

In a proxy re-encryption scheme, a semi-trusted proxy is given special power that allows it to transform a ciphertext for Alice into a ciphertext for Bob without learning any information about the messages encrypted under either key. When a proxy re-encryption scheme is constructed in an identity-based setting, it means that a proxy converts a ciphertext encrypted under Alice’s identity into a ciphertext under Bob’s. Proxy re-encryption has become more and more popular these years due to the fact that it has many practical applications. In this paper, we present an IND-CCA2 secure identity-based proxy re-encryption scheme which has several useful properties, including, multi-use, unidirectionality, etc. Finding a unidirectional, multi-use, and CCA2-secure proxy re-encryption scheme is presented as an open problem by Green et al. Fortunately, our identity-based proxy re-encryption scheme is a solution to this problem. As a middleware for fulfilling our main goal, we also propose a new construction of identity-based encryption using random padding techniques. The security of our schemes is based on the standard decisional bilinear Diffie-Hellman assumption in the random oracle model.     

基于联盟区块链的医疗健康数据安全模型

传统医疗信息系统中总是存在医疗健康数据安全存储难和共享难的问题,不同身份的人员在访问和共享医疗健康数据时都受到比较严格的限制,且验证身份和数据的真实性需要大量的资源和时间.针对传统医疗信息系统中存在的存储集中、共享安全性低和达成一致困难等问题,提出了一个基于联盟区块链的医疗健康数据安全模型.该模型根据目前的医疗资源分布...     

一种基于分布式开放资源管理服务的“云存储”(ppStore)方案研究

云存储技术因其能为用户提供安全、海量、随时随地的数据存储功能而得以快速发展.提出利用网络上现存的自然分布于世界各地的大量免费存储服务,如网盘、ftp、email以及其他形式等存储空间,将这些服务集中起来,为用户提供价廉、可靠的存储系统.介绍一种低开销、快速度、高可用、可扩展的云存储拓扑系统,以期在该拓扑架构下买现以文件...     

基于代理的移动云服务访问机制的研究与实现

移动设备的计算能力、存储能力、网络带宽、电池续航等都非常受限,而且现有的云服务基本没有针对移动设备的上述特点做优化,严重降低了移动设备访问云服务的质量。针对存在的问题,提出了面向移动设备的数据传输格式优化模型和服务Mashup模型,设计并实现了一个基于代理的移动云服务访问框架,使得移动设备通过代理服务器来访问云服务。在4种不同情况下对移动设备的服务请求响应时间、响应结果处理时间、请求响应流量和能耗分别进行了测试,结果表明实现的框架具有一定的实际意义。     

高性能的云存储安全网关设计与实现

随着互联网应用的迅速发展,数据量呈指数级增长趋势。越来越多的用户选择将数据存储备份到云端,但同时数据安全隐患也越发突出。用户对海量数据的存储备份和安全性需求越来越高。为了方便用户将本地数据迁移到云存储服务端,简化远程云端资源的访问,提供隐私保护、数据安全等附加价值,设计并实现了一个安全、高性能、大容量的混合云存储安全网关。该云存储安全网关能够提供安全、高效的数据存储备份服务,克服传统存储备份服务的不足。具有灵活的数据机密性和用户隐私管理、安全的数据存储和传输等特点。