Sheng Cao

Prof. Sheng Cao


Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
2 November 2021 in IEEE Transactions on Dependable and Secure Computing
Verified
Journal Article
Closed access
XZXiaojun Zhang
Xiaojun Zhang
CHChao Huang
Chao Huang
YZYuan Zhang
Yuan Zhang
SCSheng Cao
Sheng Cao
In this paper, we analyze the inherent characteristic of smart grid systems, where we observe that a smart meter always generates different types of electricity consumption data for one user, and a control center (CC) always conducts an in-depth statistic analysis on these data for subsequent services. Among these data, some of them are very sensitive and should be prevented for any leakage and modification. Furthermore, due to the large number of users in a smart grid system, it is advantageous for CC to receive and process the data from different users simultaneously. To this end, we propose a verifiable privacy-preserving multi-type data aggregation scheme (VPMDA) for smart grids. VPMDA enables an aggregator gateway (AG) to aggregate encrypted multi-type data and forward the aggregated data to CC, such that CC checks the integrity of aggregated data and obtains the statistic analysis results (e.g., average, variance) on the aggregated data without learning each individual data content. We further extend VPMDA to improve the performance of verifying data integrity on CC significantly. We formally prove the security of VPMDA against various attacks. We also implement a prototype of VPMDA and conduct a comprehensive performance evaluation to demonstrate its feasibility and efficiency.

1 April 2021 in Computer Communications
Verified
Journal Article
Closed access
SCSheng Cao
Sheng Cao
SDSixuan Dang
Sixuan Dang
YZYuan Zhang
Yuan Zhang
WWWei Wang
Wei Wang
NCNan Cheng
Nan Cheng

2 March 2023 in IEEE Transactions on Dependable and Secure Computing
Verified
Journal Article
Closed access
KHKe Huang
Ke Huang
YMYi Mu
Yi Mu
FRFatemeh Rezaeibagha
Fatemeh Rezaeibagha
XZXiaosong Zhang
Xiaosong Zhang
XLXiong Li
Xiong Li
SCSheng Cao
Sheng Cao
Monero is a privacy-centric cryptocurrency that allows users to obscure their transactions with multiple input and output addresses. Current research on Monero mainly focuses on identifying design vulnerabilities or optimizing towards stronger privacy, security, etc. For example, improving the design of ring confidential transaction (RingCT) protocol proposed by Noether et al. As revealed by Ali et al. in USENIX 2016, new blockchains have inadequate nodes and network computing resources to resist powerful attack (e.g. 51% attack). Obviously, Monero blockchain is not an exception. Ateniese et al. proposed the notion of redactable blockchain in EuroS&P 2017, which begins the trend of formalizing blockchain with extra cryptographic primitives. The motivation is to turn an immutable blockchain into a mutable ledger by adapting the blockchain design and integrating with new cryptographic schemes. In such a setting, users could use their private keys to perform the secure multi-party computation to reverse blockchain history. The idea of redactable blockchain has attracted many researchers to pursuit this topic. However, few works have considered the privacy-preserving setting. Even fewer have practised their designs in an actual cryptocurrency. In this paper, we seek to adapt the RingCT protocol with several building blocks. Our proposal achieves most of the desired properties for blockchain redaction. It allows multiple tracing authorities to collaboratively trace users' identities, and a system manager to perform multi-grained (including block-level, transaction-level, accumulator-level and commitment-level) redaction on block contents. Our proposal can be seen as an extension of RingCT protocol. We give rigorous security requirements and comprehensive analysis of our scheme. The performance evaluation suggested that our scheme suffers from some unscalabilities in large-scale implementations. A more elegant design to achieve stronger security and ideal scalability is deemed as a challenging and interesting future work.

17 September 2021 in International Journal of Intelligent Systems
Verified
Journal Article
Closed access
PDF file
SDSixuan Dang
Sixuan Dang
SCSheng Cao
Sheng Cao
JLJingwei Li
Jingwei Li
XZXiaosong Zhang
Xiaosong Zhang

30 July 2021 in Journal of Computer Science and Technology
Verified
Journal Article
Closed access
PDF file
HGHao Guan
Hao Guan
Yongsheng RaoYongsheng Rao
Yongsheng Rao
JZJing-Zhong Zhang
Jing-Zhong Zhang
SCSheng Cao
Sheng Cao
Xiaolin QinXiaolin Qin
Xiaolin Qin

16 September 2025 in IEEE Transactions on Network and Service Management
Verified
Journal Article
Closed access
YWYan Wang
Yan Wang
SCSheng Cao
Sheng Cao
JLJingwei Li
Jingwei Li
XZXiaosong Zhang
Xiaosong Zhang
Lately, hybrid network intrusion detection systems (HNIDSs) have progressed significantly. Through the cascade or ensemble of multiple machine learning models, HNIDS benefits from each model and achieves better performance. A widely adopted framework for designing HNIDS consists of two models: a misuse detector and an anomaly detector. However, (1) benign traffic must be analyzed by both models, reducing inference speed; (2) the misuse detector performs dual functionalities, leading to suboptimal accuracy; (3) deploying the misuse and anomaly detectors on two devices introduces substantial latency and restricts distributed deployment. In this paper, we propose a unified framework called AUF. To solve (1), we deploy the anomaly detector in the first stage rather than the second, which improves inference speed. To solve (2), we employ two independent models to implement the misuse detector’s functionality, enhancing overall accuracy. To solve (3), we ensure that the different models operate independently, supporting distributed deployment. To demonstrate the effectiveness of the AUF framework, we implement XGBoost for detection and classification and propose an adaptive k-nearest neighborhood-based approach to achieve accurate discrimination. We also introduce zero-shot learning to showcase the framework’s customized model. Extensive experiments validate the effectiveness of the AUF framework and methods. Our code is available at https://github.com/wangyann2000/A-Unified-Framework-for-Hybrid-Network-Intrusion-Detection.