报告题目：A Bayesian Approach for Spatial-temporal Traffic Modeling in Mobile Cellular Network

报告人：王学丽 教授 公司

时间：2021年5月26日（周三）14:00—15:00

地点：良乡数统楼311会议室，

腾讯会议同步直播（会议 ID：451 398 246）

报告摘要:  In this talk, we first quantify the interactive pattern between two time series: the number of users (NoU) representing user’s activity (UA) and downlink traffic load (DTL) generated from the base station (BS). We propose a Bayesian hierarchical traffic model (BHTM) for predicting DTL based on UA characteristics. Finally, the K-means clustering algorithm is used to characterize the hidden spatial-temporal association pattern. The results show that

1) there is a strong linear interaction between UA and DTL;

2) the proposed BHTM outperforms random lasso (Rlasso) and time lagged feedforward network in predicting the DTL under several evaluation rules, root mean squared error (RMSE), maximum absolute error (MAE), symmetric mean absolute percentage error (SMAPE) and hit rate;

3) when the parameters estimated from BHTM are fed into clustering analysis, the results of classification well match the reference scenario information, with the scenario recognition accuracy of 75%. Accurate traffic recognition will lead to more efficient resource management and better quality-of-service provision.

报告人简介:

王学丽，公司教授，博士生导师。现任中国现场统计研究会高维数据分会理事、计算统计分会理事、人工智能+食品安全专家委员会委员等。博士毕业于于北京大学数学科学学院，曾在美国华盛顿大员工物统计系任访问学者。在Statistic Sinica、Journal of Statistic Planning and Inference、Environmental Research and Public Health、Food Control、Environmental Science and Pollution Research等国际知名学术期刊上发表论文四十余篇。主持或完成国家重点研发计划-子课题、国家自然科学基金、全国统计科学研究项目十余项。课题论文荣获第九届全国统计科学研究优秀成果一等奖。指导的研究生多人次荣获国家奖学金和北京市优秀毕业生称号。