SARSA Based Access Control with Approximation by TileCoding
- School of Computer Science and Technology, Soochow University
Shizi Street No.1 158 Box, Suzhou, Jiangsu 215006, China
zhufei@suda.edu.cn, 20185227062@stu.suda.edu.cn,quanliu@suda.edu.cn - Provincial Key Laboratory for Computer Information Processing Technology, Soochow University
Shizi Street No.1 158 Box, Suzhou, Jiangsu 215006, China - School of Computer Science and Engineering, Changshu Institute of Technology
Changshu, Jiangsu, 215500, China
yuchenfu@cslg.edu.cn - School of Computer Science, University of Wisconsin
Eau Claire, Wisconsin, USA
zhongs@uwec.edu
Abstract
Traditional sensor nodes ignore the packet loss rate during information transmission and the access control security problem caused by server utilization when uploading data. To solve the problem, we propose a SARSA based access control method with approximation by TileCoding (SACT), which takes the sen-sor packet loss rate and the server error rate into account. The network state is estimated by the packet loss rate and variable bit error rate to get a server access control strategy to improve security performance. The eventual strategy complies with the minimum information loss and the maximum server utilization. Results of experiments show that the algorithm is capable of achieving good results in the total amount of information received by the server system. The SACT improves the server utilization rate and the overall security performance of the network.
Key words
access control, TileCoding, information security, SARSA
Digital Object Identifier (DOI)
https://doi.org/10.2298/CSIS180830022Z
Publication information
Volume 16, Issue 3 (October 2019)
Recent Advances in Information Processing and Security
Year of Publication: 2019
ISSN: 2406-1018 (Online)
Publisher: ComSIS Consortium
Full text
Available in PDF
Portable Document Format
How to cite
Zhu, F., Peng, P., Liu, Q., Fu, Y., Zhong, S.: SARSA Based Access Control with Approximation by TileCoding. Computer Science and Information Systems, Vol. 16, No. 3, 689–704. (2019), https://doi.org/10.2298/CSIS180830022Z