| In mobile environments,a lightweight client does not store the entire blockchain.When the lightweight client queries transactions from a full node,it leaks considerable information about which address or transaction is relevant to it.To protect the lightweight client’s transaction query privacy,a mobile user can run a lightweight full node to serve its lightweight client.However,the lightweight full node has the problem that its verification speed of blocks is decreasing,and this will decrease the scalability of blockchain.In addition,blockchain can be applied to manage mobile users’ access credentials which are utilized to access the public Wi-Fi hotspots.In this scenario,it should support anonymous yet accountable authentication and reduce on-chain data to improve the scalability of blockchain.Therefore,the research on privacy-preserving and scalable blockchain techniques for mobile environments has become an important research topic in the field of blockchain security.This dissertation aims to protect blockchain applications’privacy and improve blockchain applications’ scalability in mobile environments.To solve the problem of privacy leakage in querying transactions for lightweight clients,the scalability problem induced by the decreasing verification speed of blocks for lightweight full nodes,and the accountable anonymity and scalability problems when utilizing blockchain to manage mobile users’ access credentials,this dissertation studies privacy-preserving and scalable blockchain techniques for mobile environments.The main contributions of this dissertation are summarized below.(1)To solve the problem of privacy leakage in querying transactions for lightweight clients,a privacy-preserving transaction query scheme is proposed.In this scheme,an Intel SGX’s secure enclave running on a full node serves transaction query requests from lightweight clients.The proposed scheme integrates transaction index trees and oblivious RAM technologies to simultaneously support efficient transaction search and protect access pattern of externally stored blockchain data for the secure enclave.To solve the privacy leakage problem induced by the size of transaction search results,the proposed scheme designs the method of transaction query response based on the size distribution of transactions in blockchain.The proposed method can protect the size of transaction search results and decrease the communication overhead of lightweight clients.Security analysis shows that the proposed scheme can protect lightweight clients’ privacy when querying transactions from a full node.Compared with other transaction query schemes for lightweight clients,experimental results show that the proposed scheme has lower communication overhead and transaction search time.(2)To solve the problem of the decreasing verification speed of blocks for lightweight full nodes,a block verification optimization scheme,which can reduce memory requirement,is proposed.This dissertation analyzes the reason for the decreasing verification speed of blocks.For a lightweight full node that has finished blockchain synchronization,a UTXO cache preload mechanism is proposed.With the help of unconfirmed transactions,the lightweight full node calculates which unconfirmed transactions will have a greater probability of appearing in the new block,and then it pre-loads the UTXOs required for new-block verification into the UTXO cache,thereby greatly improving the verification performance.For a lightweight full node that is synchronizing blockchain,a block batch verification mechanism is proposed.With the help of the UTXO subset provided by a full node that has finished blockchain synchronization,the lightweight full node can perform block batch verification.Security analysis shows that the proposed scheme is compatible with the existing Bitcoin system and does not threaten the security of lightweight full nodes.Compared with the existing lightweight full node schemes,experimental results show that the proposed scheme can increase the verification speed of blocks and decrease the size of the UTXO cache from 450 MB to 4 MB for lightweight full nodes.(3)To solve the problem of accountable anonymity when mobile users access public Wi-Fi hotspots,an anonymous and accountable authentication scheme based on blockchain is proposed.Through cryptocurrency-based credential management technology,the proposed scheme utilizes the unmodified Bitcoin blockchain as the platform to manage and determine the ownership of access credentials without involving any trusted third party.Through verification-path-based credential exchange technology,the proposed scheme allows users to change their access credentials anonymously and supports punishment for misbehaving users.Security analysis shows that the proposed scheme can achieve users’ accountability while still maintaining users’ anonymity.Experimental results show that the proposed scheme can satisfy the authentication requirement for mobile users who access the public Wi-Fi hotspots.(4)To solve the scalability problem induced by blockchain,an anonymous and accountable authentication scheme with minimizing on-chain data is proposed.The proposed off-chain credential exchange method can reduce the number of on-chain transactions.The proposed data embedding method based on the signatures of on-chain transactions can reduce the size of on-chain transactions.Security analysis shows that the proposed scheme can still achieve users’ accountability while maintaining users’anonymity.Experimental results show that the proposed off-chain credential exchange method can reduce 99%of on-chain data,and the proposed data embedding method can at most reduce 59.67%of on-chain data. |
