移动互联网的小区流量分析

发布时间：2018-05-01 00:25

本文选题：移动互联网 + Hadoop　；参考：《北京邮电大学》2014年硕士论文

【摘要】：随着互联网的飞速发展,网络规模的不断扩展,互联网已经深入到人们的方方面面的生活中。现如今,随着移动终端的广泛普及,移动互联网已经成为人们生活中必不可少的一部分。在人们使用各种移动互联网业务的同时,在与移动网络以及网络中用户的交互过程中,产生了各式各样的网络行为。如果说前些年互联网给人们带来各种各样多姿多彩的信息,那么近几年飞速发展的移动互联网则使人们更加方便和简单地接入互联网,能够随心所欲地在任何时间地点接入互联网,更加简单快捷地获取到各种信息。本文分析了移动互联网用户行为分析的意义,介绍了移动互联网的逻辑架构以及移动互联网流量分析的特点等。接下来,介绍了用户行为分析的概念等,并阐述了海量数据处理的基础知识、难点与处理技巧,数据挖掘的目的、步骤和方法。接下来详细介绍了小区流量的统计方法,并且对各个统计方法所得到的结果的相对误差做了详细的对比分析,并且计算了各个方法的时间复杂度,对各个方法的实际运行时间做了统计以及对比分析。针对Gn接口数据存在小区重选的误差对小区流量统计的影响,本文首先对Gb接口数据与Gn接口数据的流基本特征做了详细的统计和比较,再根据Gb接口数据统计小区重选的情况,以及相关分析。以确定小区重选对Gn接口数据所得到的小区流量统计的影响。接下来,从多个维度对小区流量与时间相关性进行分析。并且运用数据挖掘知识,K-Means算法基于流量对小区进行聚类分析,并对分类后的各组表现出的特征进行深入分析和对比。
[Abstract]:With the rapid development of the Internet and the continuous expansion of the network scale, the Internet has penetrated into all aspects of people's lives. Nowadays, with the popularity of mobile terminals, mobile Internet has become an indispensable part of people's lives. When people use all kinds of mobile Internet services, they produce a variety of network behaviors in the process of interaction with mobile networks and users in the network. If the Internet brought people a variety of colorful information in the past few years, the rapid development of the mobile Internet in recent years has made it more convenient and simple for people to access the Internet. Access to the Internet anytime and anywhere, easier and faster access to all kinds of information. This paper analyzes the significance of mobile Internet user behavior analysis, introduces the logical structure of mobile Internet and the characteristics of mobile Internet traffic analysis. Then, the concept of user behavior analysis is introduced, and the basic knowledge, difficulties and processing skills of mass data processing, the purpose, steps and methods of data mining are expounded. Then the statistical methods of cell flow are introduced in detail, and the relative errors of the results obtained by each statistical method are compared and analyzed in detail, and the time complexity of each method is calculated. The actual running time of each method is statistically analyzed and compared. In view of the effect of the error of cell reselection in Gn interface data on cell traffic statistics, this paper firstly makes detailed statistics and comparison on the basic characteristics of Gb interface data and Gn interface data. Then according to GB interface data statistics cell reselection, as well as related analysis. The effect of cell reselection on cell traffic statistics obtained from Gn interface data is determined. Then, the correlation between cell traffic and time is analyzed from multiple dimensions. The K-Means algorithm of data mining is used to analyze the cluster of the cells based on the traffic, and the characteristics of each group after classification are analyzed and compared in depth.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.01;TN929.5

【参考文献】