Research On Efficient Integrity Auditing With Secure Deduplication In Cloud Computing

The emergence and rapid development of cloud computing and cloud storage provides an effective solution to the computation and storage problems of massive data.In order to ease the burden of local storage and protect the sensitive data from being leaked,an increasing amount of users tend to upload the encrypted data to cloud servers for storage and management.With the ever-growing amount of data being stored,the redundant data on cloud servers also increases rapidly.To eliminate redundant data and improve storage efficiency,secure deduplication technique has become a hot topic for research.However,in the era of big data,the data stored on cloud servers contains high commercial value and is vulnerable to internal or external malicious adversaries.Moreover,due to software and hardware breakdown or some uncontrollable factors,the integrity and availability of outsourced data may be compromised.Especially in the deduplication scene,if the deduplicated data is tampered,deleted or lost,it will cause huge losses for all users who possess the data.Integrity auditing technique can ensure the integrity and availability of outsourced data,so it is necessary to perform integrity auditing for the deduplicated data.However,most existing schemes for supporting data deduplication and integrity auditing are not efficient.In the integrity auditing phase of their schemes,the computation cost of third party auditors(TPA)is very high.Furthermore,the computational complexity of integrity auditing for TPA after block-level deduplication will further increase,especially batch auditing.To solve the problem of efficiency in existing schemes,we propose an efficient integrity auditing with secure deduplication scheme,which has been accepted by the International Journal of Embedded Systems(EI).The main contributions of our work are as follows:(1)Based on the polynomial homomorphic authentication tags and properties of bilinear pairing,we propose an efficient integrity auditing with secure deduplication scheme.In our scheme,upon performing integrity auditing for the data deduplicated by a mixed way of file-level and block-level deduplication,we aggregate the authentication tags generated by different users into the same form.Therefore,the TPA can verify the correctness of integrity proof generated by cloud service providers with a constant computation cost.Our scheme can also support batch auditing.Due to aggregating signatures of different users mentioned above,the auditing efficiency of TPA can have great improvement.This has been applied an invention patent.Furthermore,we use the merkle hash tree and proxy resignature techniques to extend our scheme to support dynamic data updates.In this scheme,we give different methods of processing data updates according to whether the users who sent the data update requests generate the authentication tags and whether the updated data has other legitimate owners.(2)According to the security model defined in our scheme,we demonstrate that our scheme achieves the desired security goals of data confidentiality,tag consistency,unforgeability of proof of ownership and integrity evidence.In addition,we also analyze the cost of storage,communication,and computation for the proposed scheme and related schemes.Then,we perform the efficiency evaluation for our scheme and the comparison schemes based on the PBC and Py Crypto cryptography library on the Linux system.The theoretical analysis and experimental results indicate that our scheme is highly efficient.