Searchable symmetric encryption (SSE) schemes allow a client to store encrypted data with a storage provider and retrieve corresponding documents without revealing the content or search keywords to the provider. However, achieving efficient SSE schemes often comes at the cost of statistical information leakage, including search, access and size patterns. The known solutions from fully homomorphic encryption or oblivious RAM often admit poor performances due to significant computational and communication overheads. Additionally, the demand for rich search expressiveness, such as Boolean queries, further complicates the design. In this paper, we introduce NEMO, a novel SSE achieving a good balance between efficiency, security and query expressiveness. NEMO utilizes function secret sharing (FSS) and replicated secret sharing-based multi-party computation (MPC) protocol, but is highly optimized for large database. For functionality, NEMO supports arbitrary Boolean queries and enables dynamic updates in a multi-user setting. For security, NEMO achieves minimal leakage by eliminating all search, access, and size patterns, while only allowing the leakage of Boolean formulas in queries. Regarding efficiency, we propose a new FSS for multi-point functions, effectively batching multiple distributed point functions, and an infix-to-postfix conversion algorithm for Boolean formula to reduce the communication rounds in the MPC protocol. A proof-of-concept implementation of NEMO demonstrates its efficiency, with a search latency of approximately 622 ms for a conjunction query with 8 keywords, even with a dataset exceeding 1 million documents.

The immutability of blockchains is critical for cryptocurrencies, but an imperative need arises for the redaction of on-chain data due to privacy-protecting laws like GPDR. Recently, Ateniese et al. (EuroS&P 2017) proposed an elegant solution to this problem based on chameleon hash functions, followed by many subsequent works. While these works offered a solution to the permissioned blockchain, the approaches were not efficient enough for the permissionless setting, in terms of either security (which may cause inconsistent historical transactions) or performance (only up to a few hundred nodes). In this paper, we investigate this problem and present Wolverine, a redactable permissionless blockchain. First, we present a formal redactable blockchain model, carefully considering transaction consistency. Next, towards a practical scheme, we introduce the novel concept of non-interactive chameleon hash (NITCH). NITCHs dynamically distribute a trapdoor key among a group and each party in the group can compute its partial share without communicating with others. Anyone who possesses enough shares can then find a valid hash collision. To prevent the static group from being compromised after a sufficiently long time, we provide a generic transform from NITCHs to decentralized random beacons (DRBs) and design a committee evolution protocol based on DRBs that refresh the group after every fixed interval of time. Based on NITCH and the committee evolution protocol, we construct Wolverine which offers important features such as scalability, transaction consistency, and public accountability. Finally, we demonstrate the practicality of Wolverine by giving a proof-of-concept implementation based on Bitcoin in Golang.

Searchable Symmetric Encryption (SSE) is a key tool for secure data processing. To date, most of the SSEs were studied alone, while an SSE supporting update operations over encrypted data remained a challenging problem due to various statistical attacks and multi-user environments. In this article, we propose SeUpdate , the first SSE scheme that simultaneously achieves keyword search and controlled update over encrypted data, with flexible read (search) and write (update) access control policies among multiple users. In SeUpdate , users do not need to share secret keys and a single query enables one to efficiently search all his authorized data. We formally define a security model, and prove our scheme have both forward and backward security. We note that the write permission of an SSE is realized for the first time. We further extend the basic scheme with dynamic access policy update and support of a large number of files. We also implement SeUpdate and some related work. The theoretical and experimental analyses demonstrate our scheme and its extension are practical and efficient.

Metadata plays an essential role in facilitating data organizing, finding, and understanding, but it also contains lots of sensitive information about the data and the data users, e.g., the location where a picture was taken. In practice, when sensitive data is encrypted before being uploaded to untrusted public clouds, an oblivious dilemma comes: if the metadata is totally encrypted, and its functionalities no longer exist; otherwise, sensitive information may be leaked. Hence, a secure and flexible mechanism for processing different fields (marked private or public for different scenario needs) of metadata simultaneously is desirable. We searched the literature for methods of achieving such a goal, it turned out that this was not explicitly considered or reasonably solved before. Therefore, in this paper, we investigate the problem of constructing privacy-enhancing metadata, namely, 1) flexible and tamper-resistant metadata setting, 2) owner-enabled secure search authorization with explicit metadata. Based on the concept of public key encryption with keyword search (PEKS), we propose a novel Authorized Keyword Search over Encrypted Data with Metadata scheme (MD-AKS), which firstly well addressed the above demands. We formalize the security model and prove the security of MD-AKS scheme. Our work maximizes the flexibility of metadata setting in two aspects: the associated metadata can be set as an arbitrary string and the costs of clients are independent of explicit metadata’s complexity. We implement MD-AKS, the theoretical comparison and experiment results further demonstrate the usability and scalability.

Ransomware attacks have caused serious data loss, and a frequent/good backup is the only solution to deal with such risks. Unfortunately, current research of data backup mainly focuses on improving data recovery efficiency, and rarely takes backups confidentiality and service DDoS attacks resistance into consideration, which greatly limits their usability in practice. In this work, we propose a novel data backup scheme called RAP (RAnsomware Protection scheme based on blockchain) to make up for the above deficiencies. Based on the principle of data isolation, we first formally discuss the role of gateways in data backup systems and deploy mechanism on gateways to resist DDoS attacks. RAP employs an optimized all-or-nothing transform (AONT) and offers the scheme for setup, backups upload and data recovery through secure channel. We define and analyze the security of RAP for data confidentiality and DDoS attacks resistance. We instantiate RAP by a consortium blockchain based on Ethereum and evaluate its performance. The running time of each phase in RAP is less than 1 millisecond excluding the $r$ead/$w$rite delay of typic data size and our scheme would be compatible with most existing blockchains, which further demonstrate its practicality and scalability.

In the cloud environment, where the cloud server cannot always be fully trusted, both data and query privacy should be well protected for secure data outsourcing. Searchable encryption provides a more practical solution to secure data storage while enabling efficient search queries. In this paper, four important problems of public key encryption with keyword search (PEKS), namely, a scheme without secure channels, conjunctive keyword search, (offline) outside and inside keyword guessing attack (full KGA) resistance and proof in the standard model, are considered. We provide an in-depth analysis of the reasons behind (offline) full KGA by considering two types of PEKS schemes as examples. In particular, we introduce the concept of server-aided secure channel free public key encryption with conjunctive keyword search (SA-SCF-PECKS) which can resist (offline) full KGA. Furthermore, we provide a concrete and efficient construction of SA-SCF-PECKS, and prove its security in the standard model. To the best of our knowledge, our proposal is the first PECKS scheme to address these four problems simultaneously. We compare the security and efficiency of our scheme with those of other related PECKS schemes in theoretical and practical ways. In general, compared with other schemes, our SA-SCF-PECKS scheme shows better performance in terms of security and efficiency.

Due to the large volume,various types and high security requirements of power data,its disaster tolerance and backup scheme should provide scalability and privacy protection.Blockchain is tamper-resistant,decentralized and traceable,which caters to such requirements.Therefore,this paper proposes a new blockchain-based disaster recovery and backup scheme of power data,integrating AONT,threshold secret sharing and other cryptographic technologies.The prototype system is implemented on an open-source blockchain platform,Ethereum.Experimental and Analysis results show that the scheme can ensure the consistency,tamper resistance and confidentiality of backup data.It is highly scalable,reducing the construction cost of infrastructure as well as the risk of single point of failure.