Blockchian Techniques For Cloud Computing Services

In the current big data era,terminal equiptments can not accommodate the massive data generated by users,and thus these data must be stored on cloud servers.Data security becomes one of the necessary research topics for cloud storage,in order to guarantee the quality of cloud storage services for users.And nowadays,security techniques in cloud storage have aroused the interest of many researchers.As the value of big data derived more from its multiple usages,integrity verification for big data has become an indispensable technique to protect the value of big data.Meanwhile,the public auditing scheme utilizes blockchain technique and deep reinforcement learning to verify the integrity of big data safely and efficiently.In addition,key management is an essential technique to ensure the security and privacy of users' data.Finally,a secure and efficient architecture for P2P distributed cloud storage is the foundation for big data in cloud storage.This thesis investigates remote data integrity verification,public auditing based on deep reinforcement learning,key management,and secure cloud storage architecture for P2P network.Specifically,this thesis utilizes the blockchain technique to explore a public auditing scheme without the third-party auditor,which innovatively stores hashtags of file blocks as the transaction data on the blockchain.Based on the public auditing scheme,this thesis explores a deep reinforcement learning-based public auditing scheme.Then,this thesis investigates the blockchain-based secure key management scheme without any common key material to eliminate the third-party.Last but not least,this thesis explores a blockchain-based scheme for secure P2P cloud storage based on the existing cloud storage architecture.The main contributions of this paper are summarized as follows.(1)A blockchain-based public auditing scheme is proposed,in which only two entities(i.e.,data owner and cloud service provider)who may not trust each other are involved,and the third-party auditor for auditing is removed.Specifically,data owners store the lightweight verification tags on the blockchain and generate a proof by constructing the merkle hash tree(MHT)using the hashtags to reduce the overhead of computation and communication for data integrity verification.Besides,this work is able to achieve 100%confidence of auditing theoretically,as the hashtag of each data block is utilized to build the MHT for the data integrity verification.Security analysis shows that the proposed scheme can defend against malicious entities and the 51%attack.Experimental results demonstrate the significant improvements on computation and communication.(2)A blockchain-empowered public auditing using deep reinforcement learning is proposed to improve the efficiency and security of public auditing in cloud storage.Most of the existing public auditing schemes in cloud storage adopt a static auditing strategy,which cannot effectively satisfy the dynamic environment of blockchain network.Meanwhile,it is a general tendency that utilizing machine learning to learn the parameters of various systems However,none of the existing public auditing schemes adopt machine learning methods to optimize the performance(i.e.,throughput,latency and etc.).The proposed scheme adopts the deep-deterministic-policy-gradient algorithm to maximize the throughput and minimize the network latency.The detailed security analysis shows that the proposed scheme is able to defend against malicious attacks derived from consensus mechanisms.Experimental results show that the proposed scheme outperforms the existing schemes in terms of throughput and network delay.(3)A blockchain-based secure key management scheme in mobile edge computing is proposed without using the common key material,and third-party is removed by utilizing blockchain technique.The scheme enables mobile devices to communicate safely and to move among subnetworks flexibly in mobile edge computing.Besides,the proposed scheme not only can resist single-point attacks in the network,but also can minimize overheads on key generation,key distribution and key storage.Then,a detailed quantitative security analysis based on the 51%attack in the blockchain network is provided to prove its security in the mobile environment.Experimental results show that the proposed scheme outperforms the basedlines in terms of the overheads on computation,communication and storage.(4)A blockchain-based secure architecture is proposed for distributed cloud storage in the P2P network.A genetic algorithm is conducted to improve the replica placement strategy of the existing architectures.This reduces the overheads on replica scheduling and transmission.And the proposed architecture is compared with two traditional architectures in terms of security and network transmission delay.Experimental results show that the proposed architecture outperforms the existing two architectures in terms of security and network latency.