Ke Zhang

Prof. Dr. Ke Zhang

School of Computer Science and Technology of China, University of Electronic Science and Technology, Chengdu, Sichuan, China
orcid

Research Focus & Metrics

Academic Career Length

19Years (2008-Present)

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

Representative Publications

View All Publications
4 August 2020 in IEEE Transactions on Industrial Informatics
Verified
Journal Article
Closed access
Ke ZhangKe Zhang
Ke Zhang
JLJiahuan Long
Jiahuan Long
XWXiaofen Wang
Xiaofen Wang
Hong-Ning DaiHong-Ning Dai
Hong-Ning Dai
KLKaitai Liang
Kaitai Liang
MIMuhammad Imran
Muhammad Imran
Industrial Internet of Things (IoT) has suffered from insufficient identity authentication and dynamic network topology, thereby resulting in vulnerabilities to data confidentiality. Recently, the attribute-based encryption (ABE) schemes have been regarded as a solution to ensure data transmission security and the fine-grained sharing of encrypted IoT data. However, most of existing ABE schemes that bring tremendous computational cost are not suitable for resource-constrained IoT devices. Therefore, lightweight and efficient data sharing and searching schemes suitable for IoT applications are of great importance. To this end, In this article, we propose a light searchable ABE scheme (namely LSABE). Our scheme can significantly reduce the computing cost of IoT devices with the provision of multiple-keyword searching for data users. Meanwhile, we extend the LSABE scheme to multiauthority scenarios so as to effectively generate and manage the public/secret keys in the distributed IoT environment. Finally, the experimental results demonstrate that our schemes can significantly maintain computational efficiency and save the computational cost at IoT devices, compared to other existing schemes.

1 August 2021 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
Ke ZhangKe Zhang
Ke Zhang
HYHanbo Ying
Hanbo Ying
Hong-Ning DaiHong-Ning Dai
Hong-Ning Dai
LLLin Li
Lin Li
YPYuanyuan Peng
Yuanyuan Peng
HYHongfang Yu
Hongfang Yu
Deep neural networks (DNNs) have shown great success in completing complex tasks. However, DNNs inevitably bring high computational cost and storage consumption due to the complexity of hierarchical structures, thereby hindering their wide deployment in Internet-of-Things (IoT) devices, which have limited computational capability and storage capacity. Therefore, it is a necessity to investigate the technologies to compact DNNs. Despite tremendous advances in compacting DNNs, few surveys summarize compacting-DNNs technologies, especially for IoT applications. Hence, this article presents a comprehensive study on compacting-DNNs technologies. We categorize compacting-DNNs technologies into three major types: 1) network model compression; 2) knowledge distillation (KD); and 3) modification of network structures. We also elaborate on the diversity of these approaches and make side-by-side comparisons. Moreover, we discuss the applications of compacted DNNs in various IoT applications and outline future directions.

18 September 2020 in Energies
Verified
Journal Article
Open access
PDF file
Ke ZhangKe Zhang
Ke Zhang
ZHZhi Hu
Zhi Hu
YZYufei Zhan
Yufei Zhan
Xiaofen WangXiaofen Wang
Xiaofen Wang
KGKeyi Guo
Keyi Guo
The smart grid is vulnerable to network attacks, thus requiring a high detection rate and fast detection speed for intrusion detection systems. With a fast training speed and a strong model generalization ability, the extreme learning machine (ELM) perfectly meets the needs of intrusion detection of the smart grid. In this paper, the ELM is applied to the field of smart grid intrusion detection. Aiming at the problem that the randomness of input weights and hidden layer bias in the ELM cannot guarantee the optimal performance of the ELM intrusion detection model, a genetic algorithm (GA)-ELM algorithm based on a genetic algorithm (GA) is proposed. GA is used to optimize the input weight and hidden layer bias of the ELM. Firstly, the input weight and hidden layer bias of the ELM are mapped to the chromosome vector of a GA, and the test error of the ELM model is set as the fitness function of the GA. Then, the parameters of the ELM intrusion detection model are optimized by genetic operation; the input weight and bias, corresponding to the minimum test error, are selected to improve the performance of the ELM model. Compared with the ELM and online sequential extreme learning machine (OS-ELM), the GA-ELM effectively improves the accuracy, detection rate and precision of intrusion detection and reduces the false positive rate and missing report rate.

23 March 2020 in IEEE Access
Verified
Journal Article
Open access
Ke ZhangKe Zhang
Ke Zhang
ZWZeyang Wang
Zeyang Wang
LGLele Guo
Lele Guo
YPYuanyuan Peng
Yuanyuan Peng
Zhi ZhengZhi Zheng
Zhi Zheng
The time interval of the observational data changes irregularly because of the difference of sensors' sampling rate, the communication delay and the target leaving observation region of the sensor sometimes. These problems of asynchronous observation data greatly reduce the tracking accuracy of the multi-sensors system. Therefore, asynchronous data fusion system is more practical than synchronous data fusion system, and worthier of study. By establishing an asynchronous track fusion model with irregular time interval of observation data and combining with the Track Quality with Multiple Model (TQMM), an asynchronous track fusion algorithm with information feedback is proposed, and the TQMM is used for weight allocation to improve the performance of the asynchronous multi-sensor fusion system. The simulation result shows that the algorithm has better tracking performance compared with other algorithms, so that this kind of problem of track-to-track fusion for asynchronous sensors is solved effectively.

15 June 2021 in IEEE Transactions on Industrial Informatics
Verified
Journal Article
Closed access
Ke ZhangKe Zhang
Ke Zhang
HCHao Chen
Hao Chen
Hong-Ning DaiHong-Ning Dai
Hong-Ning Dai
HLHongbo Liu
Hongbo Liu
ZLZhongrui Lin
Zhongrui Lin
The site selection of sports facilities is a pivotal link in the construction of city livable environment and the development of sports business in digital-twinning cities. Recent years have witnessed data mining and visualization technologies bringing the convenience as well as opportunities for intelligent site selection. However, the lack of effective and reliable systematic analysis leads to difficulties in developing sports facilities planning schemes and constructing the site-selection system. In this article, we design Sport facility Visual analysis system (SpoVis), an interactive visual analysis system for planning sports facilities as well as site selection. SpoVis provides users with the distribution status and statistical analysis of various sports facilities. Based on a comprehensive consideration of city population distribution, construction cost, existing sports facilities, traffic situation, and development potential, SpoVis provides users with a reasonable site-selection scheme of sports facilities from both macro and microperspectives and recommends results through topology and map. Meanwhile, based on the distribution of existing sports facilities and city influencing factors, a set of visual analysis components are designed to facilitate users to evaluate the status and information of existing sports facilities. We have carried out extensive experiments on a real platform with real-world data. The experimental results show that the proposed site-selection models and algorithms have excellent accuracy and operation efficiency.

21 August 2017 in Telecommunication Systems
Verified
Journal Article
Closed access
PDF file
WTWenyi Tang
Wenyi Tang
Ke ZhangKe Zhang
Ke Zhang
DJDingde Jiang
Dingde Jiang