机器学习方法和应用
当前位置: 首页 > 研究成果 > 机器学习方法和应用 > 正文

研究成果"A Pre-signed Response Method Based on Online Certificate Status Protocol Request Prediction" 发表于 Enterprise Information Systems 

信息来源:暂无 发布日期: 2021-10-21 浏览次数:


篇名:A Pre-signed Response Method Based on Online Certificate Status Protocol Request Prediction


作者:Chi-Hua Chen(陈志华),Genggeng Liu, Yu-Chih Wei,Zuoyong Li and Bon-Yeh Lin


来源:Enterprise Information Systems


年份:2021


DOI: https://doi.org/10.1080/17517575.2021.1986861


文章摘要:


This study proposes a pre-signed response method based on online certificate status protocol (OCSP) request prediction. A request prediction method (RPM) is proposed to analyze and predict potential volumes of certificate signing requests(CSRs) for OCSP certificates for a given time or period, so that responses to the requests can be generated and pre-signed during off-peak hours for load balancing. In our experiments, the OCSP request data in a certificate center was collected and analyzed for the evaluation of our proposed method. The results showed that the accuracy rate of our proposed method was about 96.17% in the prediction of traffic volumes of OCSP requests for the relevant times or periods. Therefore, our proposed method can be used to forecast potential traffic volumes of OCSP requests and help distribute the generation of responses to these requests more evenly between peak and off-peak hours.