摘要
通过对电信运营商提供的移动终端位置数据的分析,建立区域内人群聚集行为预测模型,以便于及时发现城市中所发生的人群异常聚集行为。分别统计分析移动基站下移动终端的接入量以及终端轨迹数据,通过马尔科夫链构建人群密度预测模型,最后通过对人群聚集行为的分析最终建立人群聚集行为预测模型。采用该模型对西安市大雁塔区域的人群进行预测。实验结果表明,使用马尔科夫链可以预测未来某时刻基站覆盖区域下的人群转移数量以及人群密度,结合所建立模型从而预测人群聚集行为的发生,预测精度为86.1%。
Through the analysis of the mobile terminal localtion data provided by the telecom operators,the prediction model of crowd aggregation behavior in the region is established so as to find out the crowd aggregation behavior in the city in time. The traffic volume and terminal trajectory data of the mobile terminal under the mobile base station are separately analyzed. The population density prediction model is constructed by Markov chain. Finally,the population aggregation behavior prediction model is finally established by analyzing the crowd aggregation behavior. The model is used to predict the population of the Big Wild Goose Pagoda in Xi’an. The experimental results show that the Markov chain can be used to predict the number of population transfer and population density under the coverage area of the base station at a certain moment in the future. Combined with the established model to predict the occurrence of crowd gathering behavior,the prediction accuracy is 86.1%.
作者
张成才
王瑞刚
ZHANG Chengcai;WANG Ruigang(Department of Computer Science&Technology,Xi'an University of Post&Telecommunications,Xi'an 710061)
出处
《计算机与数字工程》
2020年第3期613-616,共4页
Computer & Digital Engineering
基金
陕西省重点研发计划项目(编号:2016KTTSGY01-01)
西安邮电大学教学改革研究项目(编号:JGZ201615)资助。
关键词
人群异常聚集行为
移动基站
接入量
马尔科夫链
abnormal crowding behavior
mobile base station
access volume
Markov chain
