移动通信网络环境下的用户运动模式挖掘

Mobility pattern mining in cellular network environment

为解决移动用户运动规律获取问题,提出了一种在移动通信网络(GSM)环境下,从用户低层基站位置信息中提取出运动模式的方法.针对基站连接的振荡、重叠等问题,该方法对原始基站位置数据进行分段、窗口化、分组和聚类处理,并将原始基站位置数据表示为聚类序列.采用关联规则挖掘算法从聚类序列中提取出用户运动模式.最后,基于真实基站位置数据,通过运动路径划分实验、位置数据聚类实验和运动模式挖掘实验对系统的有效性进行了评测,并给出了系统参数设置的建议.

In order to obtain the mobility regularity of mobile users,an approach is proposed for extracting users＇ mobility patterns from cell-based location data under cellular system,e.g.GSM（global system for mobile communications）.To counter the oscillation and overlapping problem of cellular network connectivity,the raw cell-based location data is processed through segmentation,windowization,grouping and clustering,and expressed as sequences of clusters.Then,mobility patterns are extracted from the sequences of clusters based on the association rule mining algorithm.A series of experiments including route segmenting experiment,location data clustering experiment and mobility patterns mining experiment are conducted based on real-world cell phone log data to evaluate the proposed system and the suggestion of system parameters configuration is given.