As more sensitive data moves to the cloud,the need for searchable encryption grows.While PEKS allows users to search over encrypted data using public keys,a major technical pain point is its vulnerability to ciphertext distinguishing attacks.In standard PEKS, the static relationship between a keyword ciphertext and the underlying encrypted file allows a curious server to infer patterns, effectively breaking the semantic security of the system. This “leakage” makes it possible for service providers to build profiles of user behavior without ever decrypting the actual content.
To bridge this gap, the research team developed the DSPE framework. DSPE moves beyond static encryption by introducing a dynamic layer that hides the association between searchable tags and their corresponding files. This ensures “indistinguishability” even against an active server. Furthermore, DSPE introduces a critical management feature: dynamic deletion. Unlike previous models where searchable tags could only be added, DSPE allows the data receiver to instruct the server to remove specific ciphertexts, enhancing both privacy and storage management.
Experimental results confirm that DSPE maintains a high level of efficiency, with computational overhead remaining practical for modern communication platforms. By providing a framework that is both semantically secure and dynamically manageable, this research offers a robust solution for next-generation privacy-preserving cloud services.
