基于大数据的移动用户行为分析研究

Research on Mobile User Behavior Analysis Based on Big Data

基于移动大数据,本文对用户通话到达情况(次数)进行了研究。由于人们生活习惯具有时间规律性,齐次的泊松过程不能完全适合基站的通话到达规律,因此本文引入非齐次泊松过程,在一定程度上解决了齐次泊松过程的不足。根据用户通话到达情况,提出采用非齐次复合泊松过程来估算一段时间某一基站通话总共的时间,从而根据通话流量为基站的合理调度、分配资源提供依据。

Based on mobile big data,this paper studies the call arrival condition(times)of users.Due to the time regularity of people’s living habits,the homogeneous poisson process is not completely suitable for the call arrival rule of the base station.Therefore,this paper introduces the non-homogeneous poisson process to solve the deficiency of the homogeneous poisson process to some extent.According to the arrival condition of the user’s call,the non-homogeneous compound poisson process is proposed to estimate the total time of a base station’s call in a certain period of time,so as to provide a basis for the reasonable scheduling and resource allocation of the base station according to the call flow.