Study On Stateless Public Audit For Secure Cloud Storage

As one of cloud computing core services,cloud storage provides a way to solve problems of big data storage and management.However,a lot of secure events have emerged in recent years,which incurs a risk that the users do not trust the cloud service providers.The main reason is that the separation of data ownership and management,which incurs that the users cannot use a traditional way to verify data integrity.To address the problems,the technology of cloud data security audit is provided.Different from existing work,third-party is devoted to the research of stateless public audit technology,so that any third party auditor can complete efficient audit without maintaining relevant state information.This paper has done the following works:(1)In the existing public audit works of dynamic data integrity,both audit efficiency and statelessness are difficult to achieve simultaneously.In other words,existing solutions either support stateless verification in public audits,but often have large computational overhead and communication cost,or let a third-party auditor to maintain some metadata to reduce computational and communication cost,but they lose stateless verification.In view of this,a new dynamic data auditing scheme is proposed,which introduces a new entity to store a small amount of metadata.With such way,it makes third-party auditors stateless and reduces the computational and communication overhead during auditing and data updating phases.We give the strict proofs of the scheme security.In addition,performance analyses and experimental results show that the proposed scheme is more efficient in communication and computation than existing schemes.(2)Multi-version data storage in cloud storage is a common method of data storage and management.In Multi-version data storage,the data are stored with delta organized by a chain or a tree.With such way,it can save storage space and be easy to roll back the new version to the prior version.In order to realize the stateless and efficient auditing of multi-version data,a cloud data public auditing scheme based on multi-version Merkle hash tree is proposed.In order to achieve the stateless and efficient auditing of multi-version data,a cloud data public auditing scheme based on multi-version Merkle hash tree is proposed.The scheme designs a multi-version Merkle hash tree as a data authentication structure to support dynamic update of multiversion data and to reduce the storage overhead of metadata.In addition,the scheme aggregates all version information of the data block into one tag,so that the integrity of the corresponding all versions can be verified with a single data block tag,which can improve the verification performance.The rigorous security proofs of the scheme have been given,and the proposed scheme can effectively resist a series of attacks such as forgery attacks,collusion attacks etc.In addition,theoretical and experimental analyses demonstrate that the proposed scheme has less communication overhead and computation cost than existing schemes.(3)In order to meet the needs of data protection in the vehicular cloud environment,a secure and efficient auditing scheme for vehicular network data is proposed.In order to achieve lightweight auditing,especially to reduce the computing overhead of the tenant(i.e.the user)side,this scheme designs a new form of data block tags instead of the traditional complex tags,and constructs the Merkle hash tree with the hash value of the data block tags as the leaf nodes to ensure that the data block tags are unforgeable.In addition,the constructed Merkle hash tree can also support the update of data block tags,thus the scheme can achieve the reliable verification of dynamic data integrity.Finally,theoretical and experimental analyses show that this scheme can achieve high-efficiency vehicular network data security audit and has less computational overhead in the audit process than traditional methods.