Cloud traceability hash based user behavior predication mechanism for shared public cloud

Abstract

Cloud storage facilitates to share the resources with large number of users with low marginal cost. The person who is uploading the file (split into unit unit-blocks) becomes the original user and all other existing users can access the files using their private key. The existing users can also perform modification actions and update using resigning key. The integrity of the unit-blocks can be verified by public audit mechanism. Due to security reasons, an user can be revoked form the cloud (like Google drive, iCloud, drop box etc...). The unit unit-blocks which were signed by the revoked user cannot be subjected to the auditing process as it is. It has to be again resigned by the existing user which involves more computational time and utilization of resources of existing users. So, we are automating the cloud to perform these resigning operations with the existing users' resigning keys. To make this user revocation efficient, traceability of the cloud is also necessary. Using traceability results, the data owner can able to find the most frequently accessed unit-block of the cloud and the inactive member of the group. It is up to the data owner to revocate the inactive member or not. Our proposed Hash Based User Behavior Prediction Mechanism utilizing the concept of proxy resignatures, provides efficient user revocation with traceability of unit-blocks in the cloud storage.