Secure and Efficient Keyword Search Over Encrypted Graphs With Trusted Hardware
IEEE Internet of Things Journal · 01.06.2026 · 13 (11) · 23842-23861
DOI: 10.1109/jiot.2026.3671769 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Internet of Things Journal, erschienen am 01.06.2026. Verfasst von Qiuhao Wang und 6 weiteren Autor:innen. Die Arbeit nennt 48 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Basic Research Program of Shaanxi, Shaanxi Province QinChuangYuan “Scientist + Engineer” Team Building.
Abstract (Originalsprache)
Graph databases are increasingly outsourced to cloud servers for efficient management of entity relationships, yet this raises significant privacy concerns. Although PeGraph, the state-of-the-art scheme, supports encrypted conjunctive search over the graph, it suffers from limitations such as heavy client burden (requiring data users to maintain secret keys and generate extensive search tokens securely), high computational cost, result pattern leakage, and an inability to handle updates and verification. To address these issues, we propose SecGraph , a system that leverages an SGX-enabled cloud server to offload key protection and computation into a trusted enclave. We further design an SGX-aware data structure, LDCF-encoded XSet , based on a logarithmic dynamic cuckoo filter, to enable efficient plaintext computation within constrained trusted memory while mitigating pattern leakage. In addition, we introduce Twin-TSet to support dynamic graph updates while ensuring forward and backward privacy. To support verifiable search, we propose VSecGraph , which employs a procedure-oriented verification method to efficiently validate all data structures inside the enclave, avoiding costly local verification. Security analyses demonstrate the security properties of both schemes. Experimental results show that SecGraph achieves up to $208\times $ faster search than PeGraph, while VSecGraph reduces verification time by $103\times $ compared to VCQ.
Förderung
- National Natural Science Foundation of China
- Natural Science Basic Research Program of Shaanxi
- Shaanxi Province QinChuangYuan “Scientist + Engineer” Team Building
Zitieren
APA
Wang, Q., Yang, X., Qi, S., Liu, Y., Zhao, H., Li, K., & Zhao, W. (2026). Secure and Efficient Keyword Search Over Encrypted Graphs With Trusted Hardware. IEEE Internet of Things Journal, 13(11), 23842-23861. https://doi.org/10.1109/jiot.2026.3671769
Vancouver
Wang Q, Yang X, Qi S, Liu Y, Zhao H, Li K, et al. Secure and Efficient Keyword Search Over Encrypted Graphs With Trusted Hardware. IEEE Internet of Things Journal. 2026;13(11):23842-23861. doi:10.1109/jiot.2026.3671769
BibTeX
@article{Wang2026,
author = {Wang, Qiuhao and Yang, Xu and Qi, Saiyu and Liu, Yiwei and Zhao, Hongguang and Li, Ke and Zhao, Wenjia},
title = {Secure and Efficient Keyword Search Over Encrypted Graphs With Trusted Hardware},
journal = {IEEE Internet of Things Journal},
year = {2026},
volume = {13},
number = {11},
pages = {23842--23861},
doi = {10.1109/jiot.2026.3671769}
}
EUVIMED