Research Of Scalable Access Control And Deduplication For Encrypted Media Data In Cloud Environment

As a carrier that can carry abundant information,media content plays an important role in social network.With the rapid development of mobile network,people are no longer satisfied with the role of information receiver,but more interested in producing content and sharing content.Communication technology also makes people's way of recording their lives shift from traditional blogs to media and pictures,videos are more and more popular.The rapid development of media service makes people put forward new demands.On the one hand,Scalable Video Coding(SVC)technology has been widely used to improve Coding efficiency and adapt to the diversity of devices and complex network environment.It can encode video of different qualities,frame rates and resolutions into a video stream.On the other hand,as one of the most widely used media data,image storage quantity increased exponentially,have great impact to the storage performance.With the explosive growth of media data,local storage can no longer meet People's Daily needs,cloud storage and cloud computing technology came into being.Cloud storage and cloud computing technologies offer inexpensive computing resources to individuals and enterprises in a dynamically scalable manner,it also raises privacy concerns,cloud service providers are free to view the data that individuals and businesses have migrated to the cloud and may result in media data leakage.Encryption of data can guarantee the confidentiality of data,but it reduces the availability of data.The Ciphertext Policy Attribute-Based Encryption(CP-ABE)enables video encryption and fine-grained access control,but it does not achieve efficient access control of encrypted SVC video.In addition,operations such as image modification will bring a large number of similar images,and the perceived content of these images does not change.Existing methods of deduplication of encrypted data are mostly based on the binary data stream of the data to detect duplication,but they do not take into account the redundancy of similar images in the cloud environment.Therefore,this thesis focuses on the efficient access control and secure deduplication of encrypted media data,and proposes the following two schemes.Proposed an efficient access control scheme for encrypted SVC video based on CP-ABE.In order to ensure the confidentiality of media and fine-grained access control,CP-ABE mechanism is adopted to encrypt every media layer in SVC video and embed specific access policies.On this basis,a multi-level access policy structure based on secret sharing is designed to ensure that users with high quality media access rights can watch the lower quality media,which is compatible with the scalable coding characteristics of SVC video and reduces the complexity of access policy.Finally,an experimental evaluation was conducted on real data sets on mobile devices and cloud servers,the results show that the proposed scheme can protect the privacy of media and achieve fine-grained access control of media data,and at the same time,it has less computing and storage overhead.A similar deduplication scheme of encrypted images based on threshold secret sharing is proposed.Different from the traditional ciphertext deduplication scheme based on image file hashing,the proposed scheme uses the perceptual hashing algorithm to generate the hash of images,so that similar images have almost the same hash value.On this basis,a multi-server assisted deduplication algorithm based on threshold secret sharing is designed.In the image encryption upload stage,a content key generation mechanism based on blind signature is designed.The image owner interactes with multiple edge servers based on the blind signature mechanism to generate labels and content keys,and the cloud media center checks whether the labels already exist.If the number of existing labels exceeds the set threshold value,it is judged that similar pictures are no longer uploaded,and inter-server check can realize the flexible expansion of the system.In the image access phase,the image owner can use the recovered content key to complete the access of similar images.Finally,an experimental evaluation was conducted on simulation data sets,and the experimental results show that the proposed scheme can greatly reduce the required storage space.