文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > Maximizing user utility in video streaming applications
侵权投诉

Maximizing user utility in video streaming applications

Abstract—In this paper, we study some of the design tradeoffs of video streaming systems in networks with QoS guarantees. We approach this problem by using a utility function to quantify the benefit a user derives from the quality of the received video se

MaximizingUserUtilityinVideo

StreamingApplications

CarlosE.Luna,AssociateMember,IEEE,LisimachosP.Kondi,Member,IEEE,and

AggelosK.Katsaggelos,Fellow,IEEE

Abstract—Inthispaper,westudysomeofthedesigntradeoffsofvideostreamingsystemsinnetworkswithQoSguarantees.Weapproachthisproblembyusingautilityfunctiontoquantifythebenefitauserderivesfromthequalityofthereceivedvideose-quence.Wealsoconsiderthecosttothenetworkuserforstreamingthevideosequence.Wehaveformulatedthisutilitymaximizationproblemasajointconstrainedoptimizationproblemwherewemaximizethedifferencebetweentheutilityandthenetworkcost,subjecttotheconstraintthatthedecoderbufferdoesnotunder-flow.Inthismanner,wecanfindtheoptimaltradeoffbetweenvideoqualityandnetworkcost.Wepresentadeterministicdy-namicprogrammingapproachforboththeconstantbitrateandrenegotiatedconstantbitrateserviceclasses.Experimentalresultsdemonstratethebenefitsandtheperformanceoftheproposedap-proach.

IndexTerms—Qualityofservice(QoS),renegotiatedconstantbitrate(RCBR),renegotiatedservices,userutility,videostreaming.

I.INTRODUCTION

T

HEexplosivegrowthoftheWorldWideWeb(WWW)hasgeneratedrapidlyincreasingdemandforapplicationsthatallowreal-timeplaybackofdigitalvideoandaudio.Theabilitytodelivermultimediacontentforreal-timedisplay,i.e.,streaming,willplayapivotalroleinthedevelopmentofnext-generationInternetapplications.Inthispaper,weonlydiscussthestreamingofvideo,buttheideasdevelopedcanbeappliedtoothertypesofmedia.

Theprocessofstreamingvideoinvolvesaclient,whichre-questsavideostreamfromaserver.Theencodedvideostreamcanbegeneratedinrealtimebasedontherequestorbepre-en-codedandstoredinavideodatabase.Theserverisresponsibleforpacketizingtheencodedvideosequenceanddeliveringittotheclient,whilemonitoringitsdelivery.

Therearetwomajorrequirementsforthevideostreamingprocesstobesuccessful.First,thenumberoflostpacketsinthenetworkmustnotbeexcessive,otherwiseitcanresultinunac-ceptablelevelsofvideoquality.Second,thedelayexperiencedbyavideoframeasittraversesthenetworkmustbeconstantifthedisplayandencoderaretooperateatthesameframerate

ManuscriptreceivedAugust2,2000;revisedOctober1,2002.ThispaperwasrecommendedbyAssociateEditorH.Gharavi.

C.E.LunaandA.K.KatsaggelosarewiththeDepartmentofElectricalandComputerEngineering,NorthwesternUniversity,Evanston,IL60208USA(e-mail:carlos@ece.northwestern.edu;aggk@ece.northwestern.edu).

L.P.KondiiswiththeDepartmentofElectricalEngineering,StateUniversityofNewYorkatBuffalo,Amherst,NY14260USA(e-mail:lkondi@eng.buf-falo.edu).

DigitalObjectIdentifier10.1109/TCSVT.2002.808439

[1],[2].Datacontainedinpacketsthatarrivetoolatetobedis-playedisconsidereduseless.

ThepresentdayInternetisabest-effortnetwork,i.e.,allpacketsaretreatedequallyandthereisnoguaranteethatthedelayandpacketlossrequirementsmentionedabovewillbemet.Forthisreason,considerableresearchefforthasbeendevotedtothedeliveryofvideooverunreliablenetworks.Muchofthisworkfocusesonthedevelopmentoferrorconcealmenttechniquestocompensatefortheeffectsofpacketlossesinthenetwork.In[3],signalprocessingapproachestodealwithnetworklosses,whenstreamingvideoovertheInternet,arepresented.Theauthorexploressomeofthelimitationsofsuchapproachesandhighlightssomepromisingareasforresearch.Alternatively,thevideosignalisencodedinaresilientwaytonetworklosses.Suchapproachestypicallyemployastochasticmodelofthecommunicationchannelandfocusoncodingmodeselection[4],[5]oronratecontrol[6].Arecentreviewoferror-resilienttechniquesforvideocommunicationsispresentedin[7](andreferencestherein).

Inthenetworkingcommunity,therehasbeenalotofinterestindevelopingmechanismstosupportmultipleclassesofser-viceonasinglenetwork.Theseeffortshaveledtothedevel-opmentofATMnetworksand,morerecently,toproposalstosupportmultipleclassesofserviceontheInternet[8],[9].Eachoftheseclassesofservicecanbecharacterizedintermsofthethroughputanddelaythatthenetworkwillguarantee.Theseguaranteesareknownasqualityofservice(QoS)guarantees.InordertoprovideQoSguarantees,resourcessuchasbandwidthandbuffersneedtobeallocatedtoeachconnection.ThenetworkusermustselecttheappropriateserviceclassandQoSfortheirapplication.Pricingandchargingfornetworkresourceusageareimportanttoolstoencourageefficientutilizationofnetworkresources[10].Thus,videostreamingsystemsinthisenviron-menthavetobeevaluatedalongtwodimensions:receivedvideoqualityandcostinnetworkresources.

Theconstraintsimposedonavideoapplicationbythenet-workwerefirststudiedin[11].Init,theauthorsproposethejointcontrolofsourceandchannelrateinordertoachieveconsistentvideoqualityandmeettransmissionrateconstraints.In[12],theauthorconsideredamoresophisticatedapproachtojointencoderandchannelratecontrol,withthegoalofachievingconsistentvideoqualityinterms

Maximizing user utility in video streaming applications

Maximizing user utility in video streaming applications

of

第1页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email