Privacy preserving distributed point function keyword search for secure shared cloud storage

This work addresses critical challenges in secure cloud storage by introducing a novel, privacy-preserving keyword search framework over encrypted data. Unlike traditional methods that compromise either performance or privacy, this work presents a multi-client searchable encryption model that integrates Distributed Point Function (DPF) for non-interactive queries and dual-layer encryption (AES + ECC) to safeguard both data and keyword indices. A key novelty lies in the incorporation of Garbled Bloom Filters and Cuckoo Hashing for fast, collision-resistant keyword indexing, coupled with segmented search distribution to significantly reduce latency. Additionally, Wegman authentication ensures integrity verification of search results, enhancing trustworthiness. The proposed system achieves a strong balance between data confidentiality, search efficiency, and storage cost through cache-enabled repeated query handling and compressed keyword structures. Experimental evaluation confirms substantial improvements in search speed, storage optimization, and data security, making it highly scalable and practical for secure shared cloud environments.