国产日韩欧美一区二区三区三州_亚洲少妇熟女av_久久久久亚洲av国产精品_波多野结衣网站一区二区_亚洲欧美色片在线91_国产亚洲精品精品国产优播av_日本一区二区三区波多野结衣 _久久国产av不卡

?

無(wú)線網(wǎng)絡(luò)鏈路傳輸調(diào)度策略研究綜述

2018-01-15 09:43鄭旭李建中趙旭
關(guān)鍵詞:鏈路無(wú)線網(wǎng)絡(luò)調(diào)度

鄭旭+李建中+趙旭

摘要: 關(guān)鍵詞: 中圖分類號(hào): 文獻(xiàn)標(biāo)志碼: A文章編號(hào): 2095-2163(2017)06-0001-05

Abstract: The link scheduling policy for wireless networks determines the set of links to transmit in each time slot. Each link refers to a pair of source and destination nodes, and all the selected links can transmit simultaneously. To gain a general understanding of the current scheduling policies can help build a proper evaluation on the performance of the whole network, and also set some basic principles for the design of the networks. This paper surveys the main branches of current scheduling policies, and gives some insights on the opening problems and their challenges. More specifically, the scheduling policies designed for maximized throughput, quality of service, and application are introduced. For the open question, the paper mainly introduces the design of applicationlevel scheduling policy.

Keywords:

0引言

無(wú)線網(wǎng)絡(luò)作為當(dāng)下重要的支持各種應(yīng)用場(chǎng)景中數(shù)據(jù)流通的媒介,持續(xù)的引起了人們的關(guān)注。包括信息物理融合系統(tǒng)、移動(dòng)社交網(wǎng)絡(luò)、手機(jī)網(wǎng)絡(luò)等多個(gè)系統(tǒng)均依賴于無(wú)線網(wǎng)絡(luò)的技術(shù)支持。然而,受到其固有特征的限制,無(wú)線網(wǎng)絡(luò)中的數(shù)據(jù)傳輸不同于一般的有線網(wǎng)絡(luò)。這一差別主要表現(xiàn)在以下幾個(gè)方面:

首先,無(wú)線網(wǎng)絡(luò)中的不同鏈路,當(dāng)其設(shè)備足夠接近時(shí)會(huì)產(chǎn)生沖突。這一情況不同于有線網(wǎng)絡(luò)各條鏈路獨(dú)立傳輸?shù)那闆r,為無(wú)線網(wǎng)絡(luò)中各鏈路的傳輸提出了更多的挑戰(zhàn)。單條鏈路只有在不存在沖突時(shí)才可以順利完成傳輸。

其次,受到環(huán)境干擾,無(wú)線網(wǎng)絡(luò)中的鏈路通常是不穩(wěn)定的,因此鏈路進(jìn)行傳輸時(shí)會(huì)出現(xiàn)傳輸失敗的情況,需要?jiǎng)討B(tài)地選擇鏈路質(zhì)量較好的鏈路進(jìn)行數(shù)據(jù)的傳輸。

基于上述原因,無(wú)線網(wǎng)絡(luò)的傳輸調(diào)度策略設(shè)計(jì)成為了一個(gè)頗具挑戰(zhàn)的研究領(lǐng)域。鏈路傳輸調(diào)度策略選擇在每個(gè)時(shí)間片斷內(nèi)進(jìn)行傳遞的鏈路集合,選定的鏈路互不沖突。目前的傳輸調(diào)度策略主要可以分為三類,具體內(nèi)容分述如下:

1)基于吞吐量的調(diào)度策略。這一類策略在每個(gè)時(shí)間段內(nèi)嘗試最大化地利用網(wǎng)絡(luò)資源,同時(shí)兼顧各個(gè)節(jié)點(diǎn)的等待隊(duì)列長(zhǎng)度。這類方法沿用了有線網(wǎng)絡(luò)中的若干思想,即在每一時(shí)刻最大化網(wǎng)絡(luò)中總的吞吐量,充分利用帶寬。

2)基于服務(wù)質(zhì)量的調(diào)度策略。不同于簡(jiǎn)單的數(shù)據(jù)交換網(wǎng)絡(luò),無(wú)線網(wǎng)絡(luò)經(jīng)常需要承載一些功能更為復(fù)雜的應(yīng)用,這些應(yīng)用對(duì)鏈路上數(shù)據(jù)的傳輸不僅有數(shù)量要求,還有服務(wù)質(zhì)量要求,如數(shù)據(jù)傳輸延時(shí)、傳輸比例和服務(wù)頻率等。調(diào)度策略在進(jìn)行鏈路選擇時(shí),需要以滿足各條鏈路的相應(yīng)服務(wù)質(zhì)量為目標(biāo)。

3)基于應(yīng)用的調(diào)度策略??紤]無(wú)線網(wǎng)絡(luò)經(jīng)常需要支持一些具體應(yīng)用,保證應(yīng)用的正常運(yùn)行便是一個(gè)順理成章的設(shè)計(jì)思路,即在選擇鏈路時(shí),優(yōu)先考慮鏈路上的應(yīng)用是否能夠正常運(yùn)行。

了解當(dāng)前調(diào)度策略的研究現(xiàn)狀能夠幫助研究人員在為新應(yīng)用場(chǎng)景設(shè)計(jì)底層無(wú)線網(wǎng)絡(luò)時(shí)提供一系列建議,包括對(duì)鏈路層協(xié)議的設(shè)計(jì),以及對(duì)整個(gè)網(wǎng)絡(luò)的性能的直觀印象,進(jìn)而決定是否需要通過(guò)額外的技術(shù)增加網(wǎng)絡(luò)帶寬。

基于上述原因,本文將總結(jié)現(xiàn)有針對(duì)無(wú)線網(wǎng)絡(luò)的主要傳輸調(diào)度策略的研究進(jìn)展,重點(diǎn)介紹各種策略所對(duì)應(yīng)的具體場(chǎng)景,以及能達(dá)到的傳輸性能。同時(shí),分析探討一種較為復(fù)雜的應(yīng)用場(chǎng)景,即視頻傳輸中的調(diào)度機(jī)制。最后,給出今后傳輸調(diào)度策略設(shè)計(jì)所面對(duì)的新挑戰(zhàn)和未解決問(wèn)題。

1以吞吐量為目標(biāo)的調(diào)度策略

在考慮無(wú)線網(wǎng)絡(luò)中的數(shù)據(jù)傳輸時(shí),研究人員首先沿用有線網(wǎng)絡(luò)中的設(shè)計(jì)思想,提出應(yīng)當(dāng)最大化網(wǎng)絡(luò)中的吞吐量,即單位時(shí)間內(nèi)網(wǎng)絡(luò)的總傳輸速率。基于上述思想,文獻(xiàn)[1]提出了一種吞吐量?jī)?yōu)化的傳輸調(diào)度策略。這一策略在一般的無(wú)線網(wǎng)絡(luò)中,考慮在每個(gè)時(shí)間槽內(nèi)調(diào)度總權(quán)重最大的互不沖突鏈路集合進(jìn)行傳輸,此處權(quán)重指各鏈路數(shù)據(jù)等待隊(duì)列長(zhǎng)度和傳輸速率的比值。吞吐量?jī)?yōu)化定義為該傳輸策略能夠支持所有被某個(gè)傳輸調(diào)度策略所指定的鏈路數(shù)據(jù)抵達(dá)速率向量,保證網(wǎng)絡(luò)中各鏈路在其抵達(dá)速率下隊(duì)列長(zhǎng)度有限。文獻(xiàn)[2]深入探究了更多的權(quán)值定義方式,例如考慮通信信道質(zhì)量、數(shù)據(jù)包延時(shí)限制等,并且證明了在這些定義方式下,權(quán)值最大化的調(diào)度策略仍然是吞吐量?jī)?yōu)化的。文獻(xiàn)[3]分析了在最大化權(quán)重調(diào)度策略中鏈路上隊(duì)列的長(zhǎng)度以及延時(shí)長(zhǎng)度。

然而,在一般的無(wú)線網(wǎng)絡(luò)及干擾模型中,尋找一組互不干擾的鏈路集,并且保證其權(quán)值之和最大是一個(gè)NP難題。因此,最大化權(quán)值的調(diào)度策略通常需要極高的時(shí)間開(kāi)銷?;谏鲜鲈颍诖_保吞吐量性能保證的前提下,一系列低時(shí)間開(kāi)銷的調(diào)度策略也隨即出現(xiàn)。一些策略采用選擇和比較類的自適應(yīng)策略[4-5]。這些策略時(shí)間開(kāi)銷較小,并且能夠在多個(gè)時(shí)間片段后逐步達(dá)到吞吐量?jī)?yōu)化的表現(xiàn)。但是這些工作無(wú)法適用于包含信道衰減和變化的情況[6]。文獻(xiàn)[7]證明了極大化權(quán)重的調(diào)度策略同樣能夠滿足容量區(qū)域中一定比例的數(shù)據(jù)抵達(dá)速率向量,這一比例的大小為極大化權(quán)值的總權(quán)重同最大權(quán)值的比值。受這一成果啟發(fā),許多后續(xù)研究將重點(diǎn)轉(zhuǎn)移至如何在所有鏈路中尋找一個(gè)更佳優(yōu)化的權(quán)值極大化獨(dú)立鏈路子集(Maximal weight independent set,MWIS)。文獻(xiàn)[8-9]又分別提出了不同的MWIS選擇策略,能夠保證所選鏈路權(quán)值的大小。而且,文獻(xiàn)[10-12]也分別針對(duì)不同的網(wǎng)絡(luò)模型提出了不同的分布式調(diào)度策略。endprint

2以鏈路服務(wù)質(zhì)量為目標(biāo)的調(diào)度策略

易見(jiàn),上述的研究工作均以如何保證更多的數(shù)據(jù)在網(wǎng)絡(luò)中流通作為傳輸調(diào)度策略設(shè)計(jì)的核心思想之一。然而,實(shí)際的無(wú)線網(wǎng)絡(luò)通常需要承載更加復(fù)雜的系統(tǒng),這些系統(tǒng)不僅僅對(duì)較長(zhǎng)時(shí)間內(nèi)的總體數(shù)據(jù)傳輸量具有要求,也對(duì)鏈路服務(wù)質(zhì)量提出了一些規(guī)范要求。以針對(duì)環(huán)境監(jiān)測(cè)的無(wú)線傳感器網(wǎng)絡(luò)為例,在完成數(shù)據(jù)收集的同時(shí),由于監(jiān)測(cè)數(shù)據(jù)的重要性,各個(gè)節(jié)點(diǎn)同樣希望其數(shù)據(jù)能夠及時(shí)地傳輸至基站,即滿足鏈路傳輸延時(shí)的限制,同時(shí),網(wǎng)絡(luò)中的簇頭節(jié)點(diǎn)需要周期性地監(jiān)控網(wǎng)絡(luò)中各個(gè)節(jié)點(diǎn)的存活狀況,進(jìn)而提出了各鏈路傳輸調(diào)度頻率的要求。針對(duì)上述情況,研究人員展開(kāi)了一系列的研究工作,并對(duì)其重點(diǎn)研究進(jìn)展可做闡釋解析如下。

文獻(xiàn)[13]提出了一種兼顧吞吐量和傳輸延時(shí)的傳輸調(diào)度策略,這一策略沿用了最大吞吐量策略的設(shè)計(jì)思路,但是引入了數(shù)據(jù)等待時(shí)間的信息,進(jìn)而在保證吞吐量?jī)?yōu)化的同時(shí),也切實(shí)保證了各數(shù)據(jù)的最壞等待時(shí)間。文獻(xiàn)[14-16]均將時(shí)間劃分為多個(gè)時(shí)間分段,每個(gè)數(shù)據(jù)包會(huì)在其到達(dá)的時(shí)間分段結(jié)束前過(guò)期。這些工作主要在于不同的應(yīng)對(duì)信號(hào)衰減的策略[17],不一致的等待時(shí)間分布[18],或者在每個(gè)時(shí)間分組的開(kāi)始階段是否能夠獲得所有數(shù)據(jù)包信息[19]等。與此同時(shí),文獻(xiàn)[20]提出了一種考慮數(shù)據(jù)包傳輸比例的調(diào)度策略,能夠維護(hù)各條鏈路在一定時(shí)間內(nèi)未能成功傳輸?shù)臄?shù)據(jù)包個(gè)數(shù),從而平衡各鏈路的數(shù)據(jù)包傳輸比例。

文獻(xiàn)[21]提出了一種針對(duì)背壓算法的數(shù)據(jù)傳輸調(diào)度策略,這一策略能夠克服數(shù)據(jù)會(huì)通過(guò)一條較長(zhǎng)路徑進(jìn)行傳輸?shù)膯?wèn)題,從而改進(jìn)了傳輸?shù)难訒r(shí)。文獻(xiàn)[22]改進(jìn)了[21]的工作,在縮小傳輸延時(shí)的前提下,證明了所提出的策略同時(shí)是吞吐量?jī)?yōu)化的。文獻(xiàn)[23-24]研究了無(wú)線網(wǎng)絡(luò)中考慮服務(wù)規(guī)律性的鏈路調(diào)度問(wèn)題。文獻(xiàn)[25]提出了一個(gè)新的參數(shù)來(lái)衡量服務(wù)規(guī)律性的表現(xiàn),即最近服務(wù)時(shí)間間隔。在其所設(shè)計(jì)的最大化權(quán)值調(diào)度策略中,鏈路得到調(diào)度的概率隨著最近服務(wù)時(shí)間的增加而增大,因而改進(jìn)了服務(wù)等待時(shí)間,即服務(wù)規(guī)律性。文獻(xiàn)[26]進(jìn)一步分析了當(dāng)數(shù)據(jù)抵達(dá)速率靠近容量區(qū)域邊界時(shí),各鏈路服務(wù)時(shí)間間隔的表現(xiàn)。

上述研究工作主要圍繞如何在全局進(jìn)行數(shù)據(jù)傳輸調(diào)度,從而保證整個(gè)系統(tǒng)的正常運(yùn)行。然而鏈路層數(shù)據(jù)傳輸調(diào)度方法同樣需要考慮一些局部的設(shè)計(jì)。例如在多跳網(wǎng)絡(luò)中,如何將數(shù)據(jù)由原節(jié)點(diǎn)在限定時(shí)間內(nèi)傳輸至目的節(jié)點(diǎn)。文獻(xiàn)[27]研究了一種帶最大傳輸延時(shí)保證的鏈路選擇及調(diào)度策略,該方法能在每一跳選擇滿足延時(shí)要求的鏈路進(jìn)行傳輸,從而通過(guò)限制單跳延時(shí)保證了總的延時(shí)。文獻(xiàn)[28]改進(jìn)了現(xiàn)有策略,提出了一種多路徑傳播的策略,能夠平衡網(wǎng)絡(luò)總的通信負(fù)載和數(shù)據(jù)按時(shí)傳輸?shù)某晒β?。文獻(xiàn)[29]提出的策略能夠避開(kāi)網(wǎng)絡(luò)中存在的節(jié)點(diǎn)空洞。而文獻(xiàn)[30-32]則主要從兼顧傳輸延時(shí)和節(jié)點(diǎn)能量消耗的角度設(shè)計(jì)傳輸策略。

3以應(yīng)用為中心的調(diào)度策略

近年來(lái),伴隨著無(wú)線網(wǎng)絡(luò)所支持的場(chǎng)景的不斷豐富,基于應(yīng)用及應(yīng)用中各個(gè)任務(wù)的鏈路調(diào)度問(wèn)題逐漸得到了研究人員的關(guān)注。這一類問(wèn)題突破了單條鏈路服務(wù)質(zhì)量或者網(wǎng)絡(luò)吞吐量的局限性,將數(shù)據(jù)傳輸調(diào)度問(wèn)題同應(yīng)用的正常流程進(jìn)行了結(jié)合,調(diào)度問(wèn)題的優(yōu)化目標(biāo)被擴(kuò)展為保證應(yīng)用的正常運(yùn)行。

文獻(xiàn)[33]提出了一種在較大時(shí)間規(guī)模上對(duì)任務(wù)實(shí)現(xiàn)調(diào)度的策略。該策略選擇擁有較好信道質(zhì)量的鏈路及其上較長(zhǎng)等待時(shí)間的任務(wù)進(jìn)行傳輸,所有被選中的任務(wù)可以傳輸其全部數(shù)據(jù)包。文獻(xiàn)[34]提出了一種改進(jìn)的最早過(guò)期數(shù)據(jù)優(yōu)先的調(diào)度策略用于視頻的傳輸。文獻(xiàn)[35]根據(jù)不同應(yīng)用的不同特性提出了一種調(diào)度方案,能夠回避短時(shí)間內(nèi)出現(xiàn)的大量通信開(kāi)銷。任務(wù)的調(diào)度問(wèn)題同樣是云計(jì)算所要斟酌處理的問(wèn)題之一[34-36],這些工作考慮各個(gè)任務(wù)的等待時(shí)間,同時(shí)平衡各個(gè)機(jī)群的負(fù)載,減少帶寬開(kāi)銷,對(duì)于無(wú)線網(wǎng)絡(luò)數(shù)據(jù)傳輸調(diào)度策略的設(shè)計(jì)具有一定的借鑒價(jià)值。文獻(xiàn)[37-39]研究了無(wú)線傳感器網(wǎng)絡(luò)中呈現(xiàn)的不同應(yīng)用的數(shù)據(jù)收集問(wèn)題,而在過(guò)程中又分別針對(duì)應(yīng)用的點(diǎn)數(shù)據(jù)和數(shù)據(jù)窗口收集問(wèn)題提出了相應(yīng)的解決方案。文獻(xiàn)[40]設(shè)計(jì)了一種近似比為2的數(shù)據(jù)傳輸策略,其時(shí)間復(fù)雜度為數(shù)據(jù)規(guī)模的多項(xiàng)式時(shí)間。

4視頻傳輸策略中的數(shù)據(jù)傳輸機(jī)制

隨著無(wú)線網(wǎng)絡(luò)技術(shù)的迅速發(fā)展和移動(dòng)設(shè)備能力的不斷增強(qiáng),在線視頻播放正迅速成為占據(jù)當(dāng)前無(wú)線帶寬的首要應(yīng)用。因此,視頻播放中的數(shù)據(jù)傳輸調(diào)度方法受到了研究人員的廣泛青睞與矚目。

現(xiàn)有的研究工作大都關(guān)注如何通過(guò)自適應(yīng)編碼技術(shù)[41-42]平衡視頻的清晰度及流暢播放情況。當(dāng)前,許多的視頻服務(wù)商為其視頻提供了多種碼率,因此需要相應(yīng)的比特率選擇算法,用于有效權(quán)衡更高的編碼率同更平穩(wěn)的視頻播放。文獻(xiàn)[43]首次提出了如何在基于HTTP協(xié)議的視頻傳輸中選擇比特率,包括信道質(zhì)量估計(jì)、視頻比特率選擇等,能夠適應(yīng)不同網(wǎng)絡(luò)環(huán)境。文獻(xiàn)[44]采取了一種更加精確的信道質(zhì)量估計(jì)方法,并且證明了在信道可以得到準(zhǔn)確估計(jì)的情況下,所提出的傳輸策略能夠使得傳輸表現(xiàn)逼近優(yōu)化狀態(tài)。文獻(xiàn)[45]僅根據(jù)緩沖區(qū)中剩余的數(shù)據(jù)量選擇視頻編碼率,文獻(xiàn)[46]同時(shí)考慮了信道質(zhì)量和緩沖區(qū)容量來(lái)選擇視頻編碼率。文獻(xiàn)[47]提出了一種開(kāi)源的測(cè)試平臺(tái),用于測(cè)試不同視頻傳輸策略的性能,并且比較了不同參數(shù)設(shè)置對(duì)視頻傳輸?shù)挠绊憽?/p>

另一部分工作研究如何在視頻傳輸中改善用戶的觀看體驗(yàn)。文獻(xiàn)[48]通過(guò)現(xiàn)有的用戶體驗(yàn)評(píng)價(jià)指標(biāo)分析了已有的視頻傳輸策略的性能,并且提出了需要面對(duì)的挑戰(zhàn)。文獻(xiàn)[49]將視頻具體內(nèi)容同網(wǎng)絡(luò)資源聯(lián)合考慮,提出了一種集中式的用戶體驗(yàn)優(yōu)化傳輸策略。文獻(xiàn)[50]通過(guò)主觀測(cè)試發(fā)現(xiàn)用戶更愿意接受漸變而非突變的視頻比特率,進(jìn)而提出了相應(yīng)的以用戶體驗(yàn)為設(shè)計(jì)目標(biāo)的視頻傳輸策略。文獻(xiàn)[51]通過(guò)在視頻傳輸過(guò)程中保證每個(gè)用戶的比特率變化區(qū)間來(lái)提供良好的用戶體驗(yàn)。文獻(xiàn)[52]提出了一種擴(kuò)展的同時(shí)適用于視頻及音頻的傳輸策略,在信道質(zhì)量小幅波動(dòng)、大幅波動(dòng)的情況下均能夠維護(hù)一定水準(zhǔn)的用戶體驗(yàn)。文獻(xiàn)[53]將自適應(yīng)編碼技術(shù)應(yīng)用于車輛網(wǎng)絡(luò),證明了相應(yīng)的策略相較于已有工作能夠改善用戶體驗(yàn)。文獻(xiàn)[54]提出了將自適應(yīng)編碼技術(shù)應(yīng)用于游戲直播的方法。endprint

5結(jié)束語(yǔ)

關(guān)于鏈路服務(wù)質(zhì)量和用戶體驗(yàn)、應(yīng)用正常運(yùn)行的關(guān)系已經(jīng)得到了一系列研究工作的關(guān)注。在這些工作中,用戶體驗(yàn)通常由服務(wù)提供商根據(jù)其對(duì)用戶的了解來(lái)進(jìn)行定義,并且通過(guò)應(yīng)用的相應(yīng)設(shè)計(jì)和針對(duì)已有數(shù)據(jù)的分析來(lái)規(guī)范獲得。這種方法在改進(jìn)用戶的基本體驗(yàn)時(shí)較為有效,因?yàn)檫@一類型的體驗(yàn)通常由單個(gè)應(yīng)用的運(yùn)行狀況而最終確定。

然而,可能同時(shí)運(yùn)行的多個(gè)應(yīng)用為提升用戶體驗(yàn)提出了新的挑戰(zhàn),僅從單個(gè)應(yīng)用的角度考慮用戶體驗(yàn)會(huì)出現(xiàn)不足。例如當(dāng)用戶暫停其所播放的視頻時(shí),可能會(huì)存在如下兩種情況:

1)視頻頻繁的緩沖,因此用戶暫停視頻等待其緩沖結(jié)束后再行播放;

2)用戶希望暫停視頻緩沖,將網(wǎng)絡(luò)帶寬留給其它更高優(yōu)先級(jí)的應(yīng)用。

兩種情況會(huì)對(duì)視頻生成完全相反的傳輸要求,服務(wù)提供商沒(méi)有辦法根據(jù)簡(jiǎn)單的傳輸比例要求滿足各場(chǎng)景。此時(shí)保證各個(gè)應(yīng)用有平均的表現(xiàn)不會(huì)為用戶提供一個(gè)優(yōu)等的全局用戶體驗(yàn),相對(duì)地,應(yīng)當(dāng)從多個(gè)應(yīng)用的綜合表現(xiàn)來(lái)改進(jìn)體驗(yàn)。這就要求對(duì)不同應(yīng)用同時(shí)要兼具深入的了解,并且能夠發(fā)現(xiàn)用戶的行為規(guī)律。

具體來(lái)說(shuō),網(wǎng)絡(luò)運(yùn)營(yíng)商需要合理地為各個(gè)應(yīng)用分配資源來(lái)最大化全局的用戶體驗(yàn),而非只是做到單個(gè)應(yīng)用的獨(dú)立維護(hù)。然而,由于網(wǎng)絡(luò)運(yùn)營(yíng)商尚無(wú)法知曉所有的任務(wù)及用戶的關(guān)注點(diǎn),獲得一個(gè)針對(duì)多應(yīng)用的優(yōu)化的資源分配方案則是一個(gè)更具挑戰(zhàn)的任務(wù)。這一過(guò)程受限于技術(shù)有待提高、商業(yè)競(jìng)爭(zhēng)以及隱私問(wèn)題。運(yùn)營(yíng)商僅能獲取數(shù)據(jù)包級(jí)的信息,并且許多服務(wù)商由于自身利益不會(huì)參與同類網(wǎng)絡(luò)運(yùn)營(yíng)商的開(kāi)放合作。此處的另一個(gè)問(wèn)題是,即使用戶在移動(dòng)端的一些簡(jiǎn)單操作同樣可能會(huì)泄露用戶的隱私信息。因此,服務(wù)提供商通常會(huì)同用戶達(dá)成一些隱私保護(hù)的協(xié)議,從而無(wú)法真正實(shí)現(xiàn)信息的共享。基于上述原因,雖然一種全局的策略能夠進(jìn)一步改進(jìn)用戶的體驗(yàn),當(dāng)前的工作仍然僅能在單個(gè)應(yīng)用層級(jí)上展開(kāi)研究。

綜上所述,如何完全獲取用戶對(duì)各個(gè)應(yīng)用的運(yùn)行期望,進(jìn)而提出一種全局化的應(yīng)用級(jí)傳輸調(diào)度策略將會(huì)成為緊隨其后、且亟待解決的研究問(wèn)題。

參考文獻(xiàn):

[1] [ZK(#〗 TASSIULAS L, EPHREMIDES A. Stability properties of constrained queueing systems and scheduling policies for maximum throughput in multihop radio networks[J]. Automatic Control, IEEE Transactions on, 1992,37(12):1936-1948.

[2] GUPTA G R, SHROFF N B. Delay analysis for wireless networks with single hop traffic and general interference constraints[J]. IEEE/ACM Transactions on Networking (TON), 2010,18(2):393- 405.

[3] BOYACI C, YE X. Optimal delay bound for maximum weight scheduling policy in wireless networks[C]//INFOCOM, 2014 Proceedings IEEE. Toronto, ON, Canada:IEEE,2014:565-573.

[4] BUI L X, SANGHAVI S,SRIKANT R. Distributed link scheduling with constant overhead[J]. Networking, IEEE/ACM Transactions on, 2006,17(5):1467-1480.

[5] NI J,TAN B, SRIKANT R. Q-CSMA: Queuelengthbased CSMA/CA algorithms for achieving maximum throughput and low delay in wireless networks[C]//INFOCOM, 2010 Proceedings IEEE. California, USA:IEEE, 2010,20(3):1-5.

[6] LIU X,CHONG E K P,SHROFF N B. A framework for opportunistic scheduling in wireless networks[J]. Computer Networks, 2003,41(4):451-474.

[7] LIN X, SHROFF N B. The impact of imperfect scheduling on crosslayer congestion control in wireless networks[J]. Networking, IEEE/ACM Transactions on, 2006,14(2):302-315.

[8] SAKAI S,TOGASAKI M,YAMAZAKI K. A note on greedy algorithms for the maximum weighted independent set problem[J]. Discrete Applied Mathematics,2003,126(2): 313-322.

[9] WAN P J, FRIEDER O,JIA X,et al. Wireless link scheduling under physical interference model[C]//INFOCOM, 2011 Proceedings IEEE. Shanghai, China:IEEE,2011,28(6):838-845.endprint

[10]BASAGNI S. Finding a maximal weighted independent set in wireless networks[J]. Telecommunication Systems,2001,18(1):155-168.

[11]HOEPMAN J H. Simple distributed weighted matchings[J]. arXiv preprint arXiv: cs/0410047,2004.

[12]JOO C,LIN X,RYU J,et al. Distributed greedy approximation to maximum weighted independent set for scheduling with fading channels[C]//Proceedings of the fourteenth ACM international symposium on Mobile ad hoc networking and computing. Bangalore, India :ACM, 2013:89-98.

[13]NEELY M J. Delaybased network utility maximization[C]// Infocom, IEEE. San Diego, California, USA :IEEE,2010,21(1):2669-2677.

[14]HOU I,KUMAR P. Utilityoptimal scheduling in time-varying wireless networks with delay constraints[C]// Proceedings of the eleventh ACM international symposium on Mobile ad hoc networking and computing. Chicago, Illinois, USA:ACM,2010:31-40.

[15]JARAMILO J J, SRIKANT R. Optimal scheduling for fair resource allocation in ad hoc networks with elastic and inelastic traffic[J]. IEEE/ACM Transactions on Networking, 2011,19(4):1125-1136.

[16]JARAMILLO J J,SRIKANT R, YING L. Scheduling for optimal rate allocation in ad hoc networks with heterogeneous delay constraints[J]. Selected Areas in Communications, IEEE Journal on, 2011,29(5):979-987.

[17]HOU I H ,KUMAR P. Scheduling heterogeneous real-time traffic over fading wireless channels[C]//INFOCOM, 2010 Proceedings IEEE. San Diego, California, USA:IEEE, 2010:1-9.

[18]KANG X,WANG W,JARAMILLO J J,et al. On the performance of largestdeficitfirst for scheduling realtime traffic in wireless networks[J]. IEEE/ACM Transactions on Networking,2016,24(1):72-84.

[19]BUI L,SRIKANT R,STOLYAR A. Novel architectures and al gorithms for delay reduction in backpressure scheduling and routing[C]//INFOCOM 2009.Rio de Janeiro, Brazil: IEEE,2009:2936-2940.

[20]LEI Ying,SHAKKOTTAI S,REDDY A, et al. On combining shortestpath and backpressure routing over multihop wireless networks[J]. IEEE/ACM Transactions on Networking, 2011,19(3):841-854.

[21]LI R, ERYILMAZ A, LI B. Throughputoptimal wireless scheduling with regulated interservice times[C]// INFOCOM, 2013 Proceedings IEEE.Turin: IEEE, 2013:2616-2624.

[22]LI B, LI R, ERYILMAZ A. Heavytrafficoptimal scheduling with regular service guarantees in wireless networks[C]// MobiHoc, 2013 Proceedings ACM. Bangalore, India:ACM, 2013:79-88.

[23]HE Tian,ABDELZAHER T,LU Chenyang, et al. Speed: A stateless protocol for realtime communication in sensor networks[C]//2013 IEEE 33rd International Conference on Distributed Computing Systems. Rhode Island:IEEE, 2003: 46-55.endprint

[24]FELEMBAN E,LEE C G,EKICI E. Mmspeed: Multipath multispeed protocol for Qos guarantee of reliability and timeliness in wireless sensor networks[J]. IEEE transactions on mobile computing, 2006,5(6):738-754.

[25]ZHAO Lei, KAN Baoqiang, XU Yongjun, et al. FTSPEED: A faulttolerant, realtime routing protocol for wireless sensor networks[C]// Wireless Communications, Networking and Mobile Computing, 2007. WiCom 2007. International Conference on. Shanghai: IEEE, 2007: 2531-2534.

[26]CHEN M,LAI C F,WANG H. Mobile multimedia sensor networks: architecture and routing[J]. EURASIP Journal on Wireless Communications and Networking, 2011,2011(1):159.

[27]LIAN J, NAIK K. Skipping technique in face routing for wireless ad hoc and sensor networks[J]. International Journal of Sensor Networks, 2008,4(1/2):92-103.

[28]CHEN M,KWON T,MAO S, et al. Reliable and energyefficient routing protocol in dense wireless sensor networks[J]. International Journal of Sensor Networks,2008, 4(1/2):104-117.

[29]CHEN Y,NASSER N,SALTI T E,et al. A multipath Qos routing protocol in wireless sensor networks[J]. International Journal of Sensor Networks, 2010, 7(4): 207-216.

[30]YANG Y, CARDEI M. Delayconstrained energyefficient routing in heterogeneous wireless sensor networks[J]. International Journal of Sensor Networks, 2010, 7(4):236-247.

[31]HAIDER M B,IMAHORI S,SUGIHARA K. Success guaranteed routing in almost Delaunay planar nets for wireless sensor com- munication[J]. International Journal of Sensor Networks, 2011, 9(2):69-75.

[32]WU H,LIN X,LIU X,et al. Applicationlevel scheduling with deadline constraints[C]//INFOCOM, 2014 Proceedings IEEE.Toronto, ON, Canada:IEEE, 2014:2436-2444.

[33]HOU I H,SINGH R. Capacity and scheduling of access points for multiple live video streams[J]. arXiv preprint arXiv:1306.2360, 2013.

[34]SHI C,JOSHI K,PANTA R K,et al. Coast: Collaborative applicationaware scheduling of lastmile cellular traffic[C]// Proceedings of the 12th annual international conference on Mobile systems, applications, and services. Bretton Woods, NH, USA:ACM, 2014: 245-258.

[35]BOUTIN E, EKANAYAKE J, LIN W, et al. Apollo: Scalable and coordinated scheduling for cloudscale computing[C]// Proceedings of the 11th USENIX conference on Operating Systems Design and Implementation. Berkeley, CA, USA:ACM, 2014:285-300.

[36]SCHWARZKOPF M,KONWINSKI A,ABDELMALEK M,et al. Omega: Flexible, scalable schedulers for large compute clusters[C]//Proceedings of the 8th ACM European Conference on Computer Systems. Prague, Czech Republic:ACM,2013:351-364.endprint

[37]VERMA A,PEDROSA L,KORUPOLU M, et al. Largescale cluster management at google with borg[C]// Proceedings of the Tenth European Conference on Computer Systems. Bordeaux, France:ACM,2015:18.

[38]TAVAKOLI A,KANSAL A,NATH S. Online sensing task optimization for shared sensors[C]//Proceedings of the 9th ACM/IEEE International Conference on Information Processing in Sensor Networks. Stockholm, Sweden: ACM, 2010:47-57.

[39]FANG X, GAO H, LI J, et al. Applicationaware data collection in wireless sensor networks[C]// INFOCOM, 2013 Proceedings IEEE.Turin, Italy:IEEE,2013:1645-1653.

[40]GAO H,F(xiàn)ANG X,LI J,et al. Data collection in multiapplication sharing wireless sensor networks[J]. IEEE Transactions on Parallel and Distributed Systems,2015, 26(2): 403-412.

[41]van DER S A S, NADERI S V. Systems and methods for performing adaptive bitrate streaming based upon the delay of each stream and the channel rate:US, 9021119[P]. 2015-04-28.

[42]NAM H,KIM K H,CALIN D, et al. Youslow: A performance analysis tool for adaptive bitrate video streaming[C]// ACM SIGCOMM Computer Communication Review.Chicago, Illinois, USA:ACM,2014,44:111-112.

[43]STOCKHAMMER T. Dynamic adaptive streaming over http: Standards and design principles[C]// Proceedings of the second annual ACM conference on Multimedia systems. San Jose, CA, USA: ACM, 2011:133-144.

[44]ZOU X K,ERMAN J,GOPALAKRISHNAN V,et al. Can accurate predictions improve video streaming in cellular networks?[C]// Proceedings of the 16th International Workshop on Mobile Computing Systems and Applications. Santa Fe, New Mexico, USA:ACM, 2015:57-62.

[45]HUANG T Y,JOHARI R,MCKEOWN N, et al. A bufferbased approach to rate adaptation: Evidence from a large video streaming service[C]// Proceedings of the 2014 ACM conference on SIGCOMM. Chicago, Illinois, USA: ACM,2014:187-198.

[46]CICCO L D,CALDARALO V,PALMISANO V, et al. Elastic: A clientside controller for dynamic adaptive streaming over http (dash)[C]// Packet Video Workshop (PV),2013 20th International. San Jose, CA:IEEE, 2013:1-8.

[47] LEDERER S,MULLER C,TIMMERER C. Dynamic adaptive streaming over http dataset[C]// Proceedings of the 3rd Multimedia Systems Conference. Chapel Hill, NC, USA:ACM, 2012:89-94.

[48]OYMAN O, SINGH S. Quality of experience for http adaptive streaming services[J]. IEEE Communications Magazine, 2012,50(4):20-27.

[49]ESSAILI A E,SCHROEDER D,STAEHLE D, et al. Qualityofexperience driven adaptive http media delivery[C]// Communications (ICC), 2013 IEEE International Conference on. Budapest:IEEE, 2013:2480-2485.endprint

[50]MOK R K,LUO X,CHAN E W W,et al. Qdash: A QoEaware dash system[C]//Proceedings of the 3rd Multimedia Systems Conference.Chapel Hill, North Carolina: ACM, 2012: 11-22.

[51]HOUDAILLE R,GOUACHE S. Shaping http adaptive streams for a better user experience[C]// Proceedings of the 3rd Multimedia Systems Conference. Chapel Hill, North Carolina:ACM, 2012:1-9.

[52]THANG T C,HO Q D,KANG J W, et al. Adaptive streaming of audiovisual content using mpeg dash[J]. IEEE Transactions on Consumer Electronics, 2012, 58(1):78-85.

[53]MLLER C, LEDERER S, TIMMERER C. An evaluation of dynamic adaptive streaming over http in vehicular environments[C]// Proceedings of the 4th Workshop on Mobile Video. Chapel Hill, North Carolina:ACM,2012:37-42.

[54]PIRES K, SIMON G. Dash in twitch: Adaptive bitrate streaming in live game streaming platforms[C]// Proceedings of the 2014 Workshop on Design, Quality and Deployment of Adaptive Video Streaming. Sydney, Australia:ACM,2014:13-18.endprint

猜你喜歡
鏈路無(wú)線網(wǎng)絡(luò)調(diào)度
水資源平衡調(diào)度在農(nóng)田水利工程中的應(yīng)用
智能四向穿梭車系統(tǒng)的應(yīng)用與調(diào)度對(duì)策研究
10kV配網(wǎng)調(diào)度運(yùn)行故障及控制對(duì)策
無(wú)線網(wǎng)絡(luò)安全漏洞及防范策略
web3.0時(shí)代,無(wú)線網(wǎng)絡(luò)安全策略研究與防范
一種IS?IS網(wǎng)絡(luò)中的鏈路異常檢測(cè)方法、系統(tǒng)、裝置、芯片
4G無(wú)線網(wǎng)絡(luò)建設(shè)項(xiàng)目的進(jìn)度管理淺析
基于熱備份提升微波站點(diǎn)傳輸穩(wěn)定性
是時(shí)候轉(zhuǎn)換到全無(wú)線網(wǎng)絡(luò)了嗎
寿光市| 镇江市| 东阿县| 蓬莱市| 运城市| 通辽市| 得荣县| 葫芦岛市| 乌海市| 罗定市| 乌兰察布市| 乌鲁木齐县| 台山市| 承德市| 潞西市| 晋中市| 佳木斯市| 中方县| 惠东县| 潮安县| 富平县| 资阳市| 江安县| 凤庆县| 淅川县| 兖州市| 汾阳市| 邵阳市| 宁德市| 宁河县| 隆安县| 巨鹿县| 霞浦县| 宝鸡市| 肥东县| 蒲城县| 双辽市| 安化县| 南充市| 榆林市| 宜都市|