Research On Damaged Data Detection And Recovery For Cloud Storage

In recent years,with the popularity of cloud storage,more and more enterprises and individuals outsource the data to the cloud for storage.They access the network and process the data which is stored in the cloud.The cloud storage service can let users to use the storage services on-demand,however,it brings a lot of security issues.On the one hand,because the cloud storage server providers are not trusted,they may maliciously conceal the lost or corrupted data stored in the cloud,or even maliciously delete the data that the user has rarely visited,to reduce the storage cost.On the other hand,data corruption in the cloud is unavoidable,so it is necessary to safely and efficiently recover these corrupted data to ensure that the data stored in the cloud is available.In view of the above two problems,our main work of this paper includes:1)Designing and implementing a scheme about damaged data detection which is based on BLS Signature.The existing cloud storage schemes' computing costs and communication costs about damaged data detection is a linear growth with the data size.Also the detection time is long,which brings a greater burden to the user.If the existing schemes are attacked,the user's privacy will be revealed.In view of these problems,our paper presents a safe and efficient scheme about data detection.The scheme establishes a two-level distributed binary tree structure for the detecting data and then generates the aggregate tags by using the aggregation properties of the BLS signature.The damaged data blocks are found by verifying the aggregate tags,which improves the detection efficiency of the damaged data block and the performance of the whole system.Finally,the experimental analysis of the proposed scheme shows that the proposed scheme is safe and efficient.Compared with the existing scheme,the storage cost is equivalent,but the computational cost and communication cost are greatly reduced.2)Designing and implementing a data recovery scheme based on LT coding.The existing cloud storage data recovery schemes require the users to encode the data before uploading the data.When the data corruption occurs,the schemes will decode the encoded data block to restore the original data.However,when the existing schemes are used to recover the data,even if to restore a data block,you must first decode the encoded data block to get all the original data,and then encoding the original data to recover a data block.The recovery efficiency is very low,and a lot of network bandwidth will be occupied.In view of this problem,our paper presents a safe and efficient data recovery scheme,which encodes the data of the cloud storage by LT code.And then we will outsource the data to the storage servers and the recovery servers which use to carry on the data recovery.The efficiency of the data recovery is enhanced.At the same time,by the coding nature of the LT code,we achieve the dynamic update of the cloud data and verify if the cloud storage server update the data or not.Finally,the experimental analysis of the proposed scheme shows that the scheme is safe and has high recovery efficiency.In summary,our paper designs a safe and effective cloud data storage,and analyzes the related work of the cloud data storage security.As the result,we design the damage data detection scheme based on LT code and the data recovery scheme based on LT code.These two schemes mentioned in our paper not only extend the functionality of cloud data storage,but also ensure the security of the mechanism.Therefore,this article has both theoretical research significance and practical value.