Efficient Searchable Encryption Scheme Supporting Fuzzy Multi-Keyword Ranking Search on Blockchain

With the continuous growth of exponential data in IoT, it is usually chosen to outsource data to the cloud server. However, cloud servers are usually provided by third parties, and there is a risk of privacy leakage. Encrypting data can ensure its security, but at the same time, it loses the retrieval function of IoT data. Searchable Encryption (SE) can achieve direct retrieval based on ciphertext data. The traditional searchable encryption scheme has the problems of imperfect function, low retrieval efficiency, inaccurate retrieval results, and centralized cloud servers being vulnerable and untrustworthy. This paper proposes an Efficient searchable encryption scheme supporting fuzzy multi-keyword ranking search on the blockchain. The blockchain and IPFS are used to store the index and encrypted files in a distributed manner respectively. The tamper resistance of the distributed ledger ensures the authenticity of the data. The data retrieval work is performed by the smart contract to ensure the reliability of the data retrieval. The Local Sensitive Hash (LSH) function is combined with the Bloom Filter (BF) to realize the fuzzy multi-keyword retrieval function. In addition, to measure the correlation between keywords and files, a new weighted statistical algorithm combining Regional Weight Score (RWS) and Term Frequency–Inverse Document Frequency (TF-IDF) is proposed to rank the search results. The balanced binary tree is introduced to establish the index structure, and the index binary tree traversal strategy suitable for this scheme is constructed to optimize the index structure and improve the retrieval efficiency. The experimental results show that the scheme is safe and effective in practical applications.