本文選題：移動互聯(lián)網(wǎng)大數(shù)據(jù) + 業(yè)務(wù)流量分析　； 參考：《北京郵電大學(xué)》2014年碩士論文

【摘要】：隨著移動互聯(lián)網(wǎng)的高速發(fā)展,以智能手機和平板電腦為代表的移 動終端更新?lián)Q代的頻率越來越高,數(shù)據(jù)業(yè)務(wù)的種類不斷增加且其流量 占比日漸增大,給電信運營商的服務(wù)水平提出了更為嚴(yán)峻的挑戰(zhàn)。與 此同時,數(shù)據(jù)業(yè)務(wù)的迅猛發(fā)展使得蜂窩網(wǎng)絡(luò)中業(yè)務(wù)、資源和計費等的 數(shù)據(jù)量日漸龐大,從而向移動互聯(lián)中的數(shù)據(jù)存儲、處理和分析提出了 更高的要求,因此,運營商面對網(wǎng)絡(luò)流量的迅速增長,亟需新的分析 工具來充分挖掘大數(shù)據(jù)中的價值,解決將流量轉(zhuǎn)變?yōu)樾б娴碾y題,最 終實現(xiàn)蜂窩網(wǎng)絡(luò)流量經(jīng)營和智能化管道等發(fā)展戰(zhàn)略。業(yè)務(wù)流量作為用戶實際業(yè)務(wù)行為的載體,能夠在一定程度上反映 用戶的行為特征和對業(yè)務(wù)的偏好規(guī)律,因此建立準(zhǔn)確可靠地業(yè)務(wù)流量 模型有助于運營商把握用戶行為特征,并依據(jù)該特征制定相應(yīng)的運營 策略,從而提高電信運營商的服務(wù)質(zhì)量。移動互聯(lián)網(wǎng)中的業(yè)務(wù)種類繁 多,數(shù)據(jù)業(yè)務(wù)種類繁多,為簡化研究的復(fù)雜性,必須依據(jù)業(yè)務(wù)協(xié)議規(guī) 范和流量特性對其進(jìn)行合理有效的分類,最終建立準(zhǔn)確的業(yè)務(wù)流量模 型。以往關(guān)于移動互聯(lián)網(wǎng)資源的研究大多都忽視業(yè)務(wù)類型對網(wǎng)絡(luò)資源 占用情況的影響,然而業(yè)務(wù)類型和資源之間卻存在非常密切的內(nèi)在聯(lián) 系。因此,本文從實測數(shù)據(jù)出發(fā),充分研究了不同業(yè)務(wù)的流量特征以 及用戶行為特征,進(jìn)一步建立了業(yè)務(wù)資源映射模型,并在此基礎(chǔ)上開 發(fā)了基于SQL Server的業(yè)務(wù)流量分析和資源優(yōu)化平臺,從而為蜂窩 網(wǎng)絡(luò)擴容和優(yōu)化提供一定的理論指導(dǎo)和技術(shù)支撐。面對大數(shù)據(jù)對移動 互聯(lián)網(wǎng)的挑戰(zhàn),本文還引入了基于Hadoop的業(yè)務(wù)流量分析系統(tǒng),通 過HDFS分布式存儲系統(tǒng)和MapReduce并行處理框架為移動互聯(lián)網(wǎng)中 的大數(shù)據(jù)分析提供了一套解決方案,從而充分挖掘業(yè)務(wù)流量大數(shù)據(jù)中 包含的價值,提高網(wǎng)絡(luò)運營效率和服務(wù)水平。
[Abstract]:With the rapid development of mobile Internet, the frequency of mobile terminals, represented by smart phones and tablets, is becoming higher and higher, and the types of data services are increasing and their traffic ratio is increasing day by day. To the service level of telecom operators put forward a more severe challenge. At the same time, with the rapid development of data services, the amount of data such as services, resources and billing in cellular networks is increasing, which leads to the storage of data in mobile interconnection. Therefore, facing the rapid growth of network traffic, operators urgently need new analytical tools to fully tap the value of big data and solve the problem of transforming traffic into benefits. The ultimate realization of cellular network traffic management and intelligent pipeline development strategy. As the carrier of the user's actual business behavior, the service flow can reflect the characteristics of the user's behavior and the law of the user's preference for the business to a certain extent. Therefore, establishing an accurate and reliable service flow model is helpful for operators to grasp the characteristics of user behavior and formulate corresponding operation strategies according to the characteristics, thus improving the service quality of telecom operators. In order to simplify the complexity of the research, it is necessary to classify the mobile Internet according to the rules of service protocol and the characteristics of traffic, in order to simplify the complexity of the research, it is necessary to classify it reasonably and effectively. Finally, an accurate business flow model is established. In the past researches on mobile Internet resources mostly ignored the influence of service types on the occupation of network resources. However there is a very close internal connection between service types and resources. Therefore, based on the measured data, the traffic characteristics and user behavior characteristics of different services are fully studied in this paper, and the service resource mapping model is further established. On this basis, a platform for traffic analysis and resource optimization based on SQL Server is developed, which provides some theoretical guidance and technical support for the expansion and optimization of cellular networks. In the face of the challenge of big data to the mobile Internet, this paper also introduces a traffic analysis system based on Hadoop. Through the HDFS-distributed storage system and MapReduce parallel processing framework, it provides a set of solutions for big data analysis in the mobile Internet, thus fully mining the value contained in the traffic big data. Improve network operation efficiency and service level.
【學(xué)位授予單位】：北京郵電大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：TP393.01;TN929.5

【參考文獻(xiàn)】

相關(guān)期刊論文 前8條

1 張興;顏志;肖靜;蔡雯琦;王文博;;無線網(wǎng)絡(luò)業(yè)務(wù)行為——特性分析與行為建模[J];北京郵電大學(xué)學(xué)報;2010年05期

2 張柘宏;;QQ短數(shù)據(jù)業(yè)務(wù)對無線資源的需求及影響[J];硅谷;2010年22期

3 王健;張耀蘭;;不同類型數(shù)據(jù)業(yè)務(wù)對無線資源的需求及影響分析[J];中國新通信;2010年01期

4 邢丹;耿玉波;;智能手機的快速發(fā)展及其對移動網(wǎng)絡(luò)的影響分析[J];郵電設(shè)計技術(shù);2010年10期

5 李志軍;楊濤;陸曉東;;基于即時通信類業(yè)務(wù)模型的CDMA無線資源分析和優(yōu)化[J];移動通信;2011年11期

6 汪丁鼎;龔追飛;潘江永;;智能手機風(fēng)暴對移動通信網(wǎng)絡(luò)的影響及應(yīng)對策略[J];移動通信;2011年21期

7 陳陸穎;叢蓉;楊潔;于華;;高速網(wǎng)絡(luò)環(huán)境下的P2P流媒體業(yè)務(wù)分析和識別方法(英文)[J];中國通信;2011年05期

8 郭景贊;王新剛;李德屹;孟照方;;移動互聯(lián)網(wǎng)業(yè)務(wù)無線特征識別及優(yōu)化方法研究[J];郵電設(shè)計技術(shù);2012年08期

，

本文編號：2060874