無線蜂窩網(wǎng)絡中的視頻測量及傳輸機制研究
發(fā)布時間:2019-05-08 06:50
【摘要】:近年來,移動終端的視頻流量發(fā)生爆炸式的增長。思科研究報告預測移動終端視頻流量將會在2020年超過電腦視頻流量,甚至會占據(jù)全網(wǎng)絡流量的82%。在LTE無線網(wǎng)絡環(huán)境中視頻流量占總流量的比率已經達到了 40%,而且這個數(shù)字還在持續(xù)增加。因此,研究用戶在移動終端上的視頻體驗具有理論和社會價值。用戶在移動終端上的用戶體驗既與視頻傳輸本身的特征有關,還受電池容量的限制。用戶觀看視頻的行為會給移動終端帶來巨大的電量消耗,直接制約著用戶的觀看時間,從而影響用戶的觀看體驗。本論文首先通過實際網(wǎng)絡測量數(shù)據(jù)分析移動終端用戶觀看視頻的行為特點和用戶體驗,在此基礎上提出兩種針對無線蜂窩網(wǎng)絡的視頻傳輸機制,這兩種機制可以降低移動終端觀看視頻的電量,提高用戶觀看體驗。本文主要內容如下:(1)根據(jù)大量的某視頻應用日志統(tǒng)計分析視頻數(shù)據(jù)。首先,對用戶觀看視頻使用的移動設備信息進行了統(tǒng)計,包括屏幕大小、分辨率、內存大小、電池容量、價格。其次,統(tǒng)計了用戶觀看視頻的行為特征,包括用戶選擇的視頻編碼速率、網(wǎng)絡類型、用戶觀看視頻長度及觀看時間、用戶的視頻完成率等。最后,對影響視頻中斷的因素進行了統(tǒng)計分析。(2)提出一種叫做"等長度能量感知視頻塊選擇機制"的視頻傳輸機制,基本思想是在設計視頻塊大小時需要考慮如下因素:無線蜂窩網(wǎng)絡的網(wǎng)卡的Tail狀態(tài)監(jiān)聽模式帶來的浪費電量;由于用戶提前離開導致的無效的數(shù)據(jù)傳輸。該機制將建立一個數(shù)學模型,該模型給出了一個視頻塊的大小的計算方法。數(shù)值結果分析表明采用這種機制下載視頻可以降低上述兩種因素導致的耗電量。(3)提出一種叫做"變長度能量感知視頻塊選擇機制"的視頻傳輸機制,基本思想是:根據(jù)實際的網(wǎng)絡測量數(shù)據(jù)發(fā)現(xiàn),用戶繼續(xù)觀看視頻的時間分布將隨著已經觀看的時間發(fā)生變化;在此基礎上,視頻塊的大小應該隨著用戶已經觀看的時間發(fā)生變化。根據(jù)這個觀察,該機制建立了一個數(shù)學模型,在給定用戶已經觀看的時間長度條件下,該模型將得出一個下載視頻塊長度的序列。本機制通過數(shù)值對比分析實驗,證明了在相同的用戶觀看行為條件下可以節(jié)約更多的電量。
[Abstract]:In recent years, video traffic in mobile terminals has exploded. The Cisco study predicts that mobile video traffic will exceed computer video traffic by 2020, or even 82 percent of network-wide traffic. The ratio of video traffic to total traffic in the LTE wireless network has reached 40 percent, and that number continues to increase. Therefore, it is of theoretical and social value to study the video experience of users on mobile terminals. The user experience on the mobile terminal is not only related to the characteristics of the video transmission itself, but also limited by the battery capacity. The behavior of watching video will bring huge power consumption to the mobile terminal, which directly restricts the viewing time of the user, and thus affects the viewing experience of the user. Firstly, this paper analyzes the behavior characteristics and user experiences of mobile end-users watching video through the actual network measurement data, and then proposes two video transmission mechanisms for wireless cellular networks, based on which, two video transmission mechanisms for wireless cellular networks are proposed. These two mechanisms can reduce the power of mobile terminals to watch video and improve the user viewing experience. The main contents of this paper are as follows: (1) Statistical analysis of video data based on a large number of video application logs. First, the mobile device information used by users to watch video is counted, including screen size, resolution, memory size, battery capacity, and price. Secondly, the behavior characteristics of users watching video are analyzed, including the video coding rate, network type, the length and time of watching video, and the completion rate of video. Finally, the factors affecting video interruption are statistically analyzed. (2) A video transmission mechanism called "equal-length energy-sensing video block selection mechanism" is proposed. The basic idea is to consider the following factors when designing the video block size: the waste of power caused by the Tail state monitoring mode of the network card in the wireless cellular network; Invalid data transfer due to early departure of the user. This mechanism will establish a mathematical model, which gives a calculation method of the size of a video block. Numerical results show that the power consumption caused by these two factors can be reduced by using this mechanism. (3) A video transmission mechanism called variable-length energy-sensing video block selection mechanism is proposed. The basic idea is as follows: according to the actual network measurement data, the distribution of time for users to continue watching video will change with the time that they have watched; On this basis, the size of the video block should change with the time the user has watched. Based on this observation, the mechanism establishes a mathematical model. Given the length of time the user has watched, the model will obtain a sequence of download video block lengths. Through numerical comparison and analysis experiments, this mechanism proves that it can save more electricity under the same user viewing behavior.
【學位授予單位】:北京交通大學
【學位級別】:碩士
【學位授予年份】:2017
【分類號】:TN929.5
本文編號:2471696
[Abstract]:In recent years, video traffic in mobile terminals has exploded. The Cisco study predicts that mobile video traffic will exceed computer video traffic by 2020, or even 82 percent of network-wide traffic. The ratio of video traffic to total traffic in the LTE wireless network has reached 40 percent, and that number continues to increase. Therefore, it is of theoretical and social value to study the video experience of users on mobile terminals. The user experience on the mobile terminal is not only related to the characteristics of the video transmission itself, but also limited by the battery capacity. The behavior of watching video will bring huge power consumption to the mobile terminal, which directly restricts the viewing time of the user, and thus affects the viewing experience of the user. Firstly, this paper analyzes the behavior characteristics and user experiences of mobile end-users watching video through the actual network measurement data, and then proposes two video transmission mechanisms for wireless cellular networks, based on which, two video transmission mechanisms for wireless cellular networks are proposed. These two mechanisms can reduce the power of mobile terminals to watch video and improve the user viewing experience. The main contents of this paper are as follows: (1) Statistical analysis of video data based on a large number of video application logs. First, the mobile device information used by users to watch video is counted, including screen size, resolution, memory size, battery capacity, and price. Secondly, the behavior characteristics of users watching video are analyzed, including the video coding rate, network type, the length and time of watching video, and the completion rate of video. Finally, the factors affecting video interruption are statistically analyzed. (2) A video transmission mechanism called "equal-length energy-sensing video block selection mechanism" is proposed. The basic idea is to consider the following factors when designing the video block size: the waste of power caused by the Tail state monitoring mode of the network card in the wireless cellular network; Invalid data transfer due to early departure of the user. This mechanism will establish a mathematical model, which gives a calculation method of the size of a video block. Numerical results show that the power consumption caused by these two factors can be reduced by using this mechanism. (3) A video transmission mechanism called variable-length energy-sensing video block selection mechanism is proposed. The basic idea is as follows: according to the actual network measurement data, the distribution of time for users to continue watching video will change with the time that they have watched; On this basis, the size of the video block should change with the time the user has watched. Based on this observation, the mechanism establishes a mathematical model. Given the length of time the user has watched, the model will obtain a sequence of download video block lengths. Through numerical comparison and analysis experiments, this mechanism proves that it can save more electricity under the same user viewing behavior.
【學位授予單位】:北京交通大學
【學位級別】:碩士
【學位授予年份】:2017
【分類號】:TN929.5
【引證文獻】
相關碩士學位論文 前1條
1 王璐;無線網(wǎng)絡的視頻測量和內容感知的帶寬分配機制研究[D];北京交通大學;2018年
,本文編號:2471696
本文鏈接:http://www.sikaile.net/kejilunwen/xinxigongchenglunwen/2471696.html
最近更新
教材專著