發(fā)布時(shí)間：2018-07-25 09:54

【摘要】：伴隨經(jīng)濟(jì)與技術(shù)的發(fā)展,網(wǎng)絡(luò)通信已經(jīng)成為國(guó)民生活的重要工具之一,需要時(shí)刻保持穩(wěn)定與安全。網(wǎng)絡(luò)優(yōu)化是實(shí)現(xiàn)這一任務(wù)的重要手段,而網(wǎng)絡(luò)優(yōu)化的前提是必須清楚掌握當(dāng)前網(wǎng)絡(luò)的基本運(yùn)行情況。本文針對(duì)傳統(tǒng)的依靠人工分析網(wǎng)絡(luò)數(shù)據(jù)所帶來(lái)的低效率性,結(jié)合網(wǎng)絡(luò)數(shù)據(jù)量巨大這一特點(diǎn),提出將大數(shù)據(jù)挖掘技術(shù)應(yīng)用到網(wǎng)絡(luò)分析過(guò)程中。首先對(duì)數(shù)據(jù)進(jìn)行預(yù)處理,然后對(duì)數(shù)據(jù)中存在的異常小區(qū)進(jìn)行檢測(cè)并去除,接著對(duì)去除異常小區(qū)后的小區(qū)網(wǎng)絡(luò)數(shù)據(jù)進(jìn)行聚類,將相近網(wǎng)絡(luò)特性的小區(qū)劃為一類。最后針對(duì)每一類的小區(qū)進(jìn)行數(shù)據(jù)分析,獲取當(dāng)前網(wǎng)絡(luò)的運(yùn)行情況并提出網(wǎng)絡(luò)優(yōu)化方案。對(duì)于異常小區(qū)的檢測(cè),采用改進(jìn)的局部異常點(diǎn)(LOF)檢測(cè)算法。該算法將LOF算法與網(wǎng)絡(luò)數(shù)據(jù)的密度分布情況相結(jié)合,通過(guò)網(wǎng)絡(luò)數(shù)據(jù)的密度分布情況確定異常點(diǎn)的個(gè)數(shù),并獲得異常點(diǎn)集D1。然后使用LOF算法來(lái)確定相同個(gè)數(shù)的異常點(diǎn)集D2。取D1與D2的交集作為最終的異常點(diǎn)集。開(kāi)源數(shù)據(jù)上的仿真證實(shí)了該算法具有較高的精確率和較低的誤報(bào)率,同時(shí)克服了LOF算法必須知道異常點(diǎn)個(gè)數(shù)這一缺點(diǎn)。在小區(qū)聚類算法上,使用了改進(jìn)的K-means聚類算法。傳統(tǒng)的K-means聚類算法具有初始聚類中心選擇隨機(jī)性及需要手動(dòng)輸入聚類個(gè)數(shù)兩大缺陷。改進(jìn)的聚類算法依照一定的規(guī)則選擇那些密度較大又相互排斥(距離較遠(yuǎn))的點(diǎn)作為初始聚類中心,同時(shí)選擇平均類間最大相似性系數(shù)(DBI)最小時(shí)的聚類中心個(gè)數(shù)作為最終的聚類個(gè)數(shù)。改進(jìn)后的算法能夠一邊優(yōu)化聚類中心,一邊確定聚類個(gè)數(shù)。開(kāi)源數(shù)據(jù)上的仿真證實(shí)了該算法準(zhǔn)確性高,收斂速度快且誤差值小。最后,針對(duì)聚類之后的每一類小區(qū)進(jìn)行網(wǎng)絡(luò)特性分析。分析網(wǎng)絡(luò)連接設(shè)備,網(wǎng)絡(luò)利用率及網(wǎng)絡(luò)掉線情況之間的關(guān)系。每一種網(wǎng)絡(luò)掉線情況下,都求出一個(gè)網(wǎng)絡(luò)可接入性裕度。并根據(jù)網(wǎng)絡(luò)的可接入性裕度提出網(wǎng)絡(luò)優(yōu)化方案,避免網(wǎng)絡(luò)過(guò)載。
[Abstract]:With the development of economy and technology, network communication has become one of the most important tools in national life. Network optimization is an important means to realize this task, and the premise of network optimization is that the basic operation of the current network must be clearly understood. In view of the low efficiency brought by the traditional manual analysis of network data and the large amount of network data, this paper puts forward the application of big data mining technology in the process of network analysis. First, the data is preprocessed, then the abnormal cells in the data are detected and removed. Then, the data of the cell network after removing the abnormal cells are clustered, and the cells with similar network characteristics are classified into a class. Finally, the data of each kind of cell is analyzed, the current network operation is obtained and the network optimization scheme is put forward. For the detection of abnormal cells, an improved local outlier (LOF) detection algorithm is adopted. The algorithm combines the LOF algorithm with the density distribution of the network data, determines the number of outliers through the density distribution of the network data, and obtains the outliers set D1. Then the LOF algorithm is used to determine the same number of outliers. The intersection of D1 and D2 is taken as the final set of outliers. The simulation on open source data shows that the algorithm has higher accuracy rate and lower false alarm rate and overcomes the shortcoming that LOF algorithm must know the number of outliers. In the cell clustering algorithm, the improved K-means clustering algorithm is used. The traditional K-means clustering algorithm has two defects: the randomness of initial clustering center selection and the need to input the number of clusters manually. The improved clustering algorithm selects the dense and mutually exclusive points as the initial clustering centers according to certain rules. At the same time, the number of cluster centers is chosen as the final clustering number when the average maximum similarity coefficient (DBI) is minimum. The improved algorithm can determine the number of clusters while optimizing the cluster center. The simulation on open source data shows that the algorithm has high accuracy, fast convergence speed and small error. Finally, the network characteristics of each cell after clustering are analyzed. The relationship between network connection equipment, network utilization rate and network drop-off is analyzed. In each case, a network accessibility margin is obtained. According to the network accessibility margin, the network optimization scheme is proposed to avoid network overload.
【學(xué)位授予單位】：南京郵電大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP311.13

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王朔;顧進(jìn)廣;;基于K值改進(jìn)的K-means算法在入侵檢測(cè)中的應(yīng)用[J];工業(yè)控制計(jì)算機(jī);2014年07期

2 曾澤林;段明秀;;基于密度的聚類算法DBSCAN的研究與實(shí)現(xiàn)[J];科技信息;2012年30期

3 仝雪姣;孟凡榮;王志曉;;對(duì)k-means初始聚類中心的優(yōu)化[J];計(jì)算機(jī)工程與設(shè)計(jì);2011年08期

4 王品;黃焱;;改進(jìn)的OPTICS算法在調(diào)制識(shí)別中的應(yīng)用[J];計(jì)算機(jī)工程與應(yīng)用;2011年16期

5 周世兵;徐振源;唐旭清;;K-means算法最佳聚類數(shù)確定方法[J];計(jì)算機(jī)應(yīng)用;2010年08期

6 韓凌波;王強(qiáng);蔣正鋒;郝志強(qiáng);;一種改進(jìn)的k-means初始聚類中心選取算法[J];計(jì)算機(jī)工程與應(yīng)用;2010年17期

7 周涓;熊忠陽(yáng);張玉芳;任芳;;基于最大最小距離法的多中心聚類算法[J];計(jì)算機(jī)應(yīng)用;2006年06期

8 榮秋生,顏君彪,郭國(guó)強(qiáng);基于DBSCAN聚類算法的研究與實(shí)現(xiàn)[J];計(jì)算機(jī)應(yīng)用;2004年04期

9 楊風(fēng)召,朱揚(yáng)勇,施伯樂(lè);IncLOF:動(dòng)態(tài)環(huán)境下局部異常的增量挖掘算法[J];計(jì)算機(jī)研究與發(fā)展;2004年03期

10 鄭 哲;移動(dòng)通信網(wǎng)絡(luò)優(yōu)化淺析[J];電子質(zhì)量;2002年04期

相關(guān)碩士學(xué)位論文 前10條

1 王傳玉;基于異常數(shù)據(jù)挖掘算法的研究[D];天津理工大學(xué);2016年

2 黨永亮;大數(shù)據(jù)分析在移動(dòng)通信網(wǎng)絡(luò)優(yōu)化中的應(yīng)用研究[D];華中師范大學(xué);2015年

3 楊陽(yáng);數(shù)據(jù)挖掘K-means聚類算法的研究[D];湖南師范大學(xué);2015年

4 耿靈;基于EPC網(wǎng)絡(luò)的社會(huì)影響力最大化問(wèn)題[D];上海交通大學(xué);2015年

5 楊棟;針對(duì)基站聚類和用戶行為分析的移動(dòng)通信網(wǎng)絡(luò)資源優(yōu)化技術(shù)研究[D];北京郵電大學(xué);2014年

6 全擁;基于eID的虛擬資產(chǎn)審計(jì)和溯源關(guān)鍵技術(shù)研究與實(shí)現(xiàn)[D];國(guó)防科學(xué)技術(shù)大學(xué);2013年

7 劉鳳芹;K-means聚類算法改進(jìn)研究[D];山東師范大學(xué);2013年

8 詹偉成;數(shù)據(jù)挖掘在移動(dòng)通信性能指標(biāo)中的應(yīng)用研究[D];上海交通大學(xué);2012年

9 李偉斌;數(shù)據(jù)挖掘在移動(dòng)網(wǎng)絡(luò)優(yōu)化中的應(yīng)用[D];北京郵電大學(xué);2010年

10 沈亮;數(shù)據(jù)挖掘在移動(dòng)通信網(wǎng)絡(luò)優(yōu)化中的應(yīng)用[D];上海交通大學(xué);2009年

，

本文編號(hào)：2143438