無線傳感器網(wǎng)絡中英文對照外文翻譯文獻

1、（文檔含英文原文和中文翻譯）中英文對照翻譯基于網(wǎng)絡共享的無線傳感網(wǎng)絡設計摘要：無線傳感器網(wǎng)絡是近年來的一種新興發(fā)展技術(shù)，它在環(huán)境監(jiān)測、農(nóng)業(yè)和公眾健康等方面有著廣泛的應用。在發(fā)展中國家，無線傳感器網(wǎng)絡技術(shù)是一種常用的技術(shù)模型。由于無線傳感網(wǎng)絡的在線監(jiān)測和高效率的網(wǎng)絡傳送，使其具有很大的發(fā)展前景，然而無線傳感網(wǎng)絡的發(fā)展仍然面臨著很大的挑戰(zhàn)。其主要挑戰(zhàn)包括傳感器的可攜性、快速性。我們首先討論了傳感器網(wǎng)絡的可行性然后描述在解決各種技術(shù)性挑戰(zhàn)時傳感器應產(chǎn)生的便攜性。我們還討論了關于孟加拉國和加利尼亞州基于無線傳感網(wǎng)絡的水質(zhì)的開發(fā)和監(jiān)測。關鍵詞：無線傳感網(wǎng)絡、在線監(jiān)測11 .簡介無線傳感器網(wǎng)絡，是計算機設

2、備和傳感器之間的橋梁，在公共衛(wèi)生、環(huán)境和農(nóng)業(yè)等領域發(fā)揮著巨大的作用。一個單一的設備應該有一個處理器，一個無線電和多個傳感器。當這些設備在一個領域部署時，傳感裝置測量這一領域的特殊環(huán)境。然后將監(jiān)測到的數(shù)據(jù)通過無線電進行傳輸，再由計算機進行數(shù)據(jù)分析。這樣，無線傳感器網(wǎng)絡可以對環(huán)境中各種變化進行詳細的觀察。無線傳感器網(wǎng)絡是能夠測量各種現(xiàn)象如在水中的污染物含量，水灌溉流量。比如，最近發(fā)生的污染涌流進中國松花江，而松花江又是飲用水的主要來源。通過測定水流量和速度，通過傳感器對江水進行實時監(jiān)測，就能夠確定污染桶的數(shù)量和流動方向。不幸的是，人們只是在資源相對豐富這個條件下做文章，無線傳感器網(wǎng)絡的潛力在很大程

3、度上仍未開發(fā)，費用對無線傳感器網(wǎng)絡是幾個主要障礙之一，阻止了其更廣闊的發(fā)展前景。許多無線傳感器網(wǎng)絡組件正在趨于便宜化（例如有關計算能力的組件），而傳感器本身仍是最昂貴的。正如在在文獻5中所指出的，成功的技術(shù)依賴于共享技術(shù)的原因是個人設備的大量花費。然而，大多數(shù)傳感器網(wǎng)絡研究是基于一個單一的擁有長期部署的用戶，模式不利于分享。該技術(shù)管理的復雜性是另一個障礙。大多數(shù)傳感器的應用，有利于這樣的共享模型。我們立足本聲明認為傳感器可能不需要在一個長時間單一位置的原因包括：（1）一些現(xiàn)象可能出現(xiàn)變化速度緩慢，因此小批量傳感器可進行可移動部署，通過測量信號，充分捕捉物理現(xiàn)象（2）可能是過于密集，因此多余的傳

4、感器可被刪除。（3）部署時間短。我們將會在第三節(jié)更詳細的討論。上述所有假定的有關傳感器都可以進行部署和再部署。然而有很多的無線傳感器網(wǎng)絡由于其實時監(jiān)測和快速的網(wǎng)絡功能可能被利用作為共享資源。具作為共同部署資源要求，需要一些高效的技術(shù)，包括對傳感器的一些挑戰(zhàn)，如便攜性，流動頻繁的傳感器內(nèi)的部署，這使我們在第四節(jié)將會有大的挑戰(zhàn)。在本文中，我們專注于作為共享的可行性設計的傳感器網(wǎng)絡。下面我們開始闡述傳感網(wǎng)絡在孟加拉國和加利福尼亞州的水質(zhì)檢測中的應用。2 .無線傳感網(wǎng)絡在水質(zhì)監(jiān)測中的應用無線傳感器網(wǎng)絡是通過把小型計算機設備連接到各式傳感器和無線電而組成的。這些設備自適應的形成特殊網(wǎng)絡（暫時的點對點網(wǎng)絡

5、），通過無線方式對所處環(huán)境進行監(jiān)測、處理。其硬件和軟件的設計非常低功耗以此達到長期在現(xiàn)場部署的目的，即此種部署在所處環(huán)境中人為干預性小。設備大小通常從四分之一個個人數(shù)據(jù)處理機到類似一個個人數(shù)據(jù)處理機的裝置那么大。在一般情況下，資源可用性和功耗與設備大小是相一致的。例如，雖然資源可用性在很大程度上取決于傳感器的功耗，但是低功率節(jié)點（通常稱為微塵）用兩節(jié)AA電池可以運行大約一二個月。傳感器網(wǎng)絡提供密集的空間和時間上的采樣。此種取樣即使是在偏遠和難以到達的地方均可采樣。因此，它是對于在時間上和空間上要求精確采集最適用的網(wǎng)絡技術(shù)。例如無線傳感網(wǎng)絡在土壤中的應用就是個很好的例子。因為土壤環(huán)境在空間上是多

6、樣性的，需要精確的時間上的采樣。對于突然發(fā)生的變化都會被精確的采樣及時記錄下來。事實上，無線傳感器網(wǎng)絡是一種低功耗的網(wǎng)絡技術(shù)，對于一些發(fā)達地區(qū)其作為一種新興技術(shù)適用性更為廣泛。止匕外，對于公共健康方面的應用極為重要。例如，參考文獻（17）闡述了人們對于水質(zhì)的極高的關注度，“對水質(zhì)的分析起初仍然是通過實驗采樣的辦法將采集到的樣本帶回實驗室進行研究?！边@種類型的數(shù)據(jù)收集和分析通常是非常耗時的且大多是不準確的，并在許多情況下，錯過了人們對于及時關注的焦點的分析。我們參與了兩項正在進行的關于地下水質(zhì)監(jiān)測的無線傳感網(wǎng)絡部署：一項系統(tǒng)是以了解孟加拉國地下水中種的含量為主。另一項系統(tǒng)是通過研究孟加拉國地下水

7、和土壤來監(jiān)測硝酸鹽的傳播。以上我們的部署都具有類似的設置。一個塔架，是由外圍箱體式的無線設備組成，這些設備在土壤中通過長導線連接到嵌入式傳感器。每個設備可以支持7個傳感器，每個塔架都有多個設備。多路塔架被部署在目的地周圍，以達到空間上垂直和水平的密集部署。這些設備將采集到的樣本以無線方式傳送給基站以供分析。該部署的基站是一個個人數(shù)據(jù)處理機類的設備，也可以是一種輕便電腦。它是通過由太陽能提供再充電的汽車電池來進行供電。為了能夠獲得外部數(shù)據(jù)，我們的基站使用Zigbee技術(shù),或在Zigbee不可用時，使用GPRS網(wǎng)絡。在孟加拉國，在恒河三角洲的幾千萬人飲用了已被種嚴重污染的地下水，如果被污染的水量一

8、直持續(xù)，由神引起的患病率和皮膚癌將大約每年分別增加兩百萬和一萬例，由神引起的癌癥的死亡率每年將會大約增加三千例。我們對于控制種在地下水中的動態(tài)變化是難以完全了解的。在與孟加拉國的工程技術(shù)大學和麻省理工學院進行合作中，我們于2006年1月在靠近達卡的一個水稻地里部署了一個傳感網(wǎng)絡，目的是為了幫助確認這個假說成立。一個完整的塔架應該包含3部分完整的傳感器（土壤濕度，溫度，碳酸鹽，鈣，硝酸，氯，氧化還原電位，氨氮，pH值），每個部署都具有不同的深度（在地面以下1,1.5,2米），在此基礎上的壓力傳感器用來監(jiān)測水的深度。在干旱地區(qū)和半干旱地區(qū)水的短缺和不斷增加的對于水資源的消耗已經(jīng)促進人們重新再利用被

9、處理過的廢水。盡管對于水資源的再利用人類收獲了很多益處，但是已被處理的廢水對于人類的健康和環(huán)境質(zhì)量仍然存在著顯而易見的危害。解決這些危害需要進行自動的分布式的觀測和控制灌溉水量，查出它所傳輸?shù)奈廴疚?，包括暫停處理的或是還未處理的污染物，膠狀污染物，藥物，有機碳，揮發(fā)性有機化合物，治病微生物，營養(yǎng)素例如氮或磷。在加利福尼亞的帕姆代爾，一個水質(zhì)再利用現(xiàn)場是為測試土壤濕度，溫度和硝酸鹽的傳感器網(wǎng)絡而被用作的試車臺。此網(wǎng)絡集合了兩個方面：第一，確保此環(huán)境正在被監(jiān)測，第二，提供對水質(zhì)控制的反饋，從而達到優(yōu)化水流量和減少化學物質(zhì)滲透到地下。這種現(xiàn)場也可以被用來在對孟加拉國進行部署前對軟件，傳感器和硬件的測

10、試。3 .傳感器共享技術(shù)對于傳感器網(wǎng)絡數(shù)據(jù)收集，即使是最小的傳感器資源，其共享也將讓許多人受益。我們相信以下三種技術(shù)方法特別適用于傳感器共享：（1）從一系列小型傳感器大范圍部署到精確仿真。（2）從密集部署到稀疏部署逐漸移動冗余傳感器（3）在一些可能的地區(qū)縮短部署周期。在這里，我們更詳細地描述這些場景，包括我們自己和別人在執(zhí)行有關的或支持的算法時的工作的調(diào)查。（1）精確仿真人類功能的移動性就是通過手動來模擬一個使用較少傳感器的密集部署區(qū)。人們可以移動一個領域的一小套傳感器，對密集空間收集數(shù)據(jù)。該技術(shù)將是只適合于可持續(xù)發(fā)展應用中，所關注的現(xiàn)象變化非常緩慢。(2)密集到稀疏部署一些傳感器網(wǎng)絡應用需要

11、一個密集映射的環(huán)境。一旦傳感器密集部署和細節(jié)的現(xiàn)象揭示，我們可以看到它可以捕獲足夠的資料較少的傳感器，從而釋放傳感器部署在其他地方。這里，我們描述適用的工作是正在進行中的傳感器網(wǎng)絡社區(qū)。(3)部署周期短有些應用程序只需要短時間部署，因而對傳感器的共享是種理想選擇。我們在孟加拉的部署是一個帶有部署周期短的應用例子。我們要收集數(shù)據(jù)，以驗證有關晝夜變化的假設，所以我們希望數(shù)天時間來對數(shù)據(jù)進行分析。4 .挑戰(zhàn)許多挑戰(zhàn)性技術(shù)的出現(xiàn)，是為了能夠快速部署和移動傳感器，主要因為迄今為止的工作主要集中在靜態(tài)的，長期運行的部署中。我們已經(jīng)有了趨于密集化的目標，降低高密度部署使之稀疏，使周期短的部署趨于平衡，我們發(fā)

12、現(xiàn)以下三個挑戰(zhàn)是最恰當?shù)摹Ｋ惴ū仨毷蔷哂腥藱C通信功能的，對于人為錯誤是可以解決的。對于系統(tǒng)故障必須迅速查明，并最大限度地通過正確的數(shù)據(jù)進行接收。最后，系統(tǒng)必須迅速做出部署。5 .結(jié)論無線傳感器網(wǎng)絡可視為一種工具，其對于可持續(xù)發(fā)展來說具有很好的潛力。如果我們視這種發(fā)展的無線傳感網(wǎng)絡技術(shù)為共享資源的話，它就可以得到技術(shù)社區(qū)的幫助。為了使無線傳感器網(wǎng)絡作為一種共享資源得到落實，我們確定了三個有希望的技術(shù)方法：精確仿真，從密集部署到稀疏部署，實施短周期部署。我們討論了我們的工作部署，這些部署已證明了這些技術(shù)，描述了我們的過去和現(xiàn)在需要做哪些工作去面對即將出現(xiàn)的重大挑戰(zhàn)。DesigningWireles

13、sSensorNetworksasaSharedResourceforSustainableDevelopmentAbstract:Wirelesssensornetworks(WSNs)arearelativelynewandrapidlydevelopingtechnology;theyhaveawiderangeofapplicationsincludingenvironmentalmonitoring,agriculture,andpublichealth.Sharedtechnologyisacommonusagemodelfortechnologyadoptionindevelop

14、ingcountries.WSNshavegreatpotentialtobeutilizedasasharedresourceduetotheiron-boardprocessingandad-hocnetworkingcapabilities,howevertheirdeploymentasasharedresourcerequiresthatthetechnicalcommunity?rstaddressseveralchallenges.Themainchallengesincludeenablingsensorportabilitythefrequentmovementofsenso

15、rswithinandbetweendeployments,andrapidlydeployablesystems-systemsthatarequickandsimpletodeploy.We?rstdiscussthefeasibilityofusingsensornet-worksasasharedresource,andthendescribeourresearchinaddressingthevarioustechnicalchallengesthatariseinenablingsuchsensorportabilityandrapiddeployment.Wealsooutlin

16、eourexperiencesindevelopinganddeployingwaterqualitymonitoringwirelesssensornetworksinBangladeshandCalifornia.Keywords:WSNs、on-boardprocessing1 IntroductionWirelessSensorNetworks(WSNs),networksofwirelesslyconnectedsensingandcomputationaldevices,holdtremendouspromiseformanyareasofdevelopmentincludingp

17、ublichealth,theenvironment,andagriculture.Asingledevicehasaprocessor,aradio,andseveralsensors.Whenanetworkofthesedevicesisdeployedina?eld,thesensingdevicesmeasureparticularaspectsoftheenvironment.Thedevicesthencommunicatethosemeasurementsbyradiotooneanotherandtomorepowerfulcomputersfordataanalysis.I

18、nthisway,WSNscanprovidedetailedobservationsofvariousphenomenathatoccurintheenvironment.WSNsarecapableofmeasuringdiversephenomenasuchascontaminantlevelsinwater,pollutantsintheair,andthe?owofwaterforirrigation.Asanexampleofapotentialapplication,considertherecentincidentofcontaminationspillingintotheSo

19、nghuariverinChina,themainsourceofdrinkingwaterformanypeople1.Determiningrateof?owandsometimesdirectionoftheriverrequirescoordinationofmultiplesamplingpoints.Sensorsperiodicallytakingsamplesatmultiplelocationsalongtherivercoulddeterminetherate,quantity,anddirectionofcontaminant?owusingthedistributeds

20、ensingandprocessingofawirelesssensornetwork.Unfortunately,thepotentialofwirelesssensornet-worksforsustainabledevelopment2remainslargelyuntappedwhiletheyaredesignedprimarilyforrelativelyresource-richapplicationcontexts.ThecostofWSNsisoneofseveralmajorbarriersthatpreventsthemfrombeingleveragedforsusta

21、inabledevelopmentapplications.ManycomponentsofWSNsarebecomingcheaper(putingpower),butthesensorsthemselvesremainthemostexpensivecomponent3.Asstatedin5,successfultechnology-basedinternationaldevelopmentprojectsrelyonsharedtechnologyduetoexcessivecostofpersonaldevices.However,mostresearchonsensornetwor

22、ksisbasedonlong-termdeploymentsownedbyasingleuser,aparadigmnotconduciveforsharing.Thecomplexityoftechnologymanagementisanotherbarrier.WeuseGrameentelecomasasuccessfulmodel4inwhichthemanagementandmaintenanceofsharedhardwareiscentralized.Weenvisionasensornetworkmuchinthesamelight.Manysensornetworkappl

23、icationsareconducivetosuchasharedmodel.Webasethisstatementontheobservationthatsensorsmaynotberequiredinasinglelocationforextendedperiodsoftimeforreasonsincluding:(1)aphenomenonofinterestmayhaveaslowrateofchange,thusasmallnumberofsensorscanbemovedwithinadeployment,emulatingthedensityrequiredtosucient

24、lycapturethephysicalphenomena,(2)theinitialdeploymentmayhavebeentoodense,thusredundantsensorscanberemoved,and(3)thedurationofthedeploymentmaybeshort.WediscussthesescenariosinmoredetailinSection3.Allofthedeploymentscenariosmentionedaboverestontheassumptionthatsensorscanbeeasilydeployedandre-deployed.

25、WhileWSNshavegreatpotentialtobeutilizedasasharedresourceduetotheiron-boardprocessingandad-hocnetworkingcapabilities,theirdeploymentasasharedresourcerequiresthatthetechnicalcommunity?rstaddressseveralchallenges,includingenablingsensorportabilitythefrequentmovementofsensorswithinandbetweendeployments,

26、andrapidlydeployablesystems-systemsthatarequickandsimpletodeploy.ThisleadsustoourmajorchallengesinSection4.Clearly,theprimaryissuesrelatedtosuccessfultechnologyadoptionarethesocial,policy,andlogisticalquestionstobeansweredinordertoenableequitableaccessandthedesignofculturallyappropriatetechnology.Ou

27、rexperience,thoughrelevant,islimitedtoourtechnicalexpertise.Thesechallengesandothersshouldbeformulatedmoreexplicitlywiththenecessarydiverseinputfromcommunities,activists,governmentsandNGOs.Inthispaperwefocusonjustifyingthetechnicalfeasibilityofdesigningsensornetworksasasharedtechnology(Section3)andd

28、escribingthetechnicalchallengesthatmustbeaddressedtoenableWSNsasasharedtechnology(Section4).WebeginbydescribingourapplicationsinwaterqualitymonitoringinBangladeshandCalifornia(Section2).2 WSNsForWaterQualityWirelesssensornetworksaremadeupofsmallcomputationaldevicesconnectedtovarioussensorsandwireles

29、sradios.Thedevicesautomaticallyandadaptivelyformad-hocnetworks(temporarypoint-to-pointnetworks)overwirelessradiostomakedecisionsbasedonmeasurementsoftheirenvironment.Thehardwareandsoftwarearedesignedtobeextremelylowpowerinordertoenablelong-termin-situdeployments,i.e.undisturbeddeploymentsthatareleft

30、intheenvironmentwithminimalhumanintervention.DevicesizescommonlyrangefromthatofaquartertoaPDA-likedevice.Ingeneral,resourceavailabilityandpowerconsumptionarecommensuratewithsize.Forexample,whileitlargelydependsonthepowerconsumptionofthesensors,thelower-powernodes(oftencalledmotes)canrunforapproximat

31、elyonemonthon2AAbatteries.Sensornetworksprovidedensespatialandtemporalsamplingeveninremoteandhardtoreachlocations.Thus,theyarebestappliedtoapplicationsthatneeddensesamplinginspaceand/ortime.Soiapplicationsareagoodexample,becausethesoienvironmentisheterogeneousacrossspace,requiringdensespatialsamplin

32、g.Abruptchangescanthenbecapturedwithahightemporalsamplingrate.ThefactthatWSNsarelowpowerandwirelessmakesthemappealingasatechnologyfordevelopingregions,butinadditionthedensesamplingiscrucialforpublichealthapplications.Forexample,17statesthatwhilewaterqualityconcernscanbeextremelycritical,“analysisiss

33、tillprimarilyconductedinalaboriousmannerbyphysicalcollectionofasamplethatisanalyzedbackinalaboratory."Thiskindofdatacollectionandanalysisistimeconsumingandmostlyundirected,andinmanyinstancesmissesthetoxineventsofinterest.WeareinvolvedwithtwoongoingWSNdeploymentsrelatedtogroundwaterquality:asyst

34、emtounderstandtheprevalenceofarsenicinBangladeshgroundwater,andasystemtomonitornitratepropagationthroughsoilsandgroundwaterinCalifornia.Bothofourdeploymentshaveasimilarsetup.Apylon10(Figure2)consistsofanenclosurehousingthesmallwirelessdeviceswhichconnecttogroupsofsensorsembeddedatmultipledepthsinthe

35、soilthroughlongwires.Eachdevicecansupport7sensorsandtherearemultipledevicesperpylon.Multiplepylonsaredeployedaroundthe?eldtoattainverticalandhorizontalspatialdensity.Thedeviceswirelesslytransmitsamplesbacktoabase-stationforanalysis(Figure1).Thebase-stationinthesedeploymentswasaPDA-classdevice.Itcoul

36、dalsobealaptop.Itispoweredbyacar-batteryrechargedusingsolarpanels.Tomakedataexternallyaccessible,ourbase-stationisconnectedusingZigbeeorwhereZigbeeisunavailable,usingaGPRS(i.e.cellular)network.InBangladesh,tensofmillionsofpeopleintheGangesDeltadrinkgroundwaterthatisdangerouslycontaminatedwitharsenic

37、.Ifconsumptionofcontaminatedwatercontinues,theprevalenceofarsenicosisandskincancerwillbeapproximately2,000,000and100,000casesperyear,respectively,andtheincidenceofdeathfromcancerinducedbyarsenicwillbeapproximately3,000casesperyear18.Afullunderstandingofthefactorscontrollingarsenicmobilizationtogroun

38、dwaterislacking.InajointcollaborationwithscientistsattheBangladeshUniversityofEngineeringandTechnologyandMIT,wedeployedasensornetworkinJanuaryof2006inarice?eldnearDhaka,Bangladeshinordertoaidinvalidatingthishypothesis.Afullpyloncontains3completesuitesofsensors(soilmoisture,temperature,carbonate,calc

39、ium,nitrate,chloride,oxidation-reductionpotential,ammonium,andpH),eachdeployedatadi?erentdepth(1,1.5,and2metersbelowground),andapressuretransduceratthebasetomonitorwaterdepth.Waterscarcityinaridandsemi-aridregionsandincreasingdemandonwatersupplieshasstimulatedinterestinthereuseoftreatedwastewater.De

40、spitethemanybene?tstoirrigatingwithreclaimedwater,thereremainbothrealandperceivedriskstohumanhealthandenvironmentalqualitystemmingfromresidualsinthetreatedwastewater.Proactivelyaddressingtheserisksrequiresautomatingthedistributedobservationandcontroloftheirrigationwaterandthe10tracepollutantsthatitc

41、onveys,includingsuspendedordissolvedsolids(TDS),colloidalsolids,pharmaceuticals,organiccarbon,volatileorganiccompounds,pathogenicmicroorganisms,andnutrientssuchasnitrogenorphosphorus.AwaterreusesiteinPalmdale,Californiaisbeingusedasatestbedforasensornetworkwithsoilmoisture,temperature,andnitratesens

42、ors.Thenetworkfocusesontwothings:?rst,ensuringthatenvironmentalregulationsarebeingmet,andsecond,providingfeedbacktoawatercontrolsysteminordertooptimizewater?owandminimizechemicalpenetrationintothesubsurface.Thissiteisalsousedtotestthesoftware,sensors,andhardwarebeforedeployinginBangladesh.3 SensorSh

43、aringTechniquesSensorsharingwillallowmanypeopletobene?tfromsensornetworkdatacollection,evenwithminimalsensorresources.Webelievethefollowingthreetechnicalapproachesareparticularlysuitedforenablingsensorsharingforsustainabledevelopment:(1)movingasmallernumberofsensorsaroundinadeploymenttoemulatedensit

44、y,(2)graduallyremovingredundantsensorsfromadeploymenttogofromdensetosparsedeployments,and(3)leveragingshorterdeploymentcycleswherepossible.Herewedescribeeachofthesescenariosingreaterdetail,includingasurveyofourownandothers'workinimplementingrelatedorsupportingalgorithms.(1) EmulatingDensityHuman

45、-enabledmobilitycanbeusedtomanuallyemulatethee?ectofadensedeploymentusingfewersensors.Peoplecanmoveasmallsetofsensorsaroundina?eldinordertocollectdataforadensespatialmapofthe?eld.Thistechniquewillbeappropriateonlyforsustainabledevelopmentapplicationsinwhichthephenomenonofinterestchangesveryslowly,on

46、theorderofdaysorlonger.(2) DensetoSparseDeployments11Somesensornetworkapplicationsrequireadensemappingoftheenvironment.Oncesensorsaredenselydeployedanddetailsofthephenomenonarerevealed,wemayseeitispossibletocapturesu?cientinformationwithfewersensors,freeingsensorsfordeploymentelsewhere.Herewedescrib

47、eapplicableworkwhichisongoinginthesensornetworkcommunity.(3) ShortDeploymentCyclesSomeapplicationsonlyrequireshort-durationdeploymentsandthereforeareidealforsensorsharing.OurdeploymentinBangladeshisanexampleofanapplicationwithashortdeploymentcycle.Wewantedtocollectdatatovalidateahypothesisaboutdiurn

48、alvariations,andsowewantedseveraldaysofdataforanalysis.4 ChallengesNumeroustechnicalchallengesariseinordertobeabletoquicklydeployandmovesensors,primarilybecausetheworktodatehaslargelyfocusedonstatic,long-runningdeployments.Giventhatwehavethegoalstoemulatedensity,reducedensedeploymentstosparseones,an

49、dleverageshortdeploymentscycles,we?ndthefollowingthreechallengestobethemostpertinent.Algorithmsmustbeinteractiveandrobusttohumanerror.Faultsinthesystemmustbequicklyidenti?edtomaximizetheamountofgooddatareceived.Finally,systemsmustbemadetoberapidlydeployable.5 ConclusionWirelesssensornetworkshavethep

50、otentialtobeausefultoolforsustainabledevelopment.Thiscanbefacilitatedbythetechnicalcommunityifwefocusonissueswithdevelopingwirelesssensornet-worksasasharedtechnology.Inorder12toimplementWSNsasasharedresource,weidenti?edthreepromisingtechnicalapproaches:emulatingdensity,movingfromdensetosparsedeploym

51、ents,andimplementingshortdeploymentcycles.Wediscussedourworkondeploymentsthathavedemonstratedthesetechniquesanddescribedourpastandongoingworktoaddressthemajorchallengeswhicharise.13References1 TheGeneralAssembly.2005WorldOutcomeDocumentref:ResolutionadoptedbytheGeneralAssembly60/1.2005WorldSummitOut

52、come.UnitedNations,October2005.2 M.Batalin,W.Kaiser,R.Pon,G.S.Sukhatme,G.Pottie,Y.Yu,J.Gordon,M.H.Rahimi,andD.Estrin.Taskallocationforevent-awarespatiotemporalsamplingofenvironmentalvariables.InIEEE/RSJIntl.Conf.onIntelligentRobotsandSystems,August2005.3 M.Batalin,M.H.Rahimi,Y.Yu,D.Liu,A.Kansal,G.S.

53、Sukhatme,W.Kaiser,M.Hansen,G.Pottie,M.Srivastava,andD.Estrin.Callandresponse:Experimentsinsamplingtheenvironment.InProcs.ofACMSenSys,pp.25W8,Nov2004.4 D.Bornstein.ThePriceofaDream.TheUniversityofChicagoPress,1997.5 E.Brewer,M.Demmer,B.Du,M.Ho,M.Kam,S.Nedevschi,J.Pal,R.Patra,S.Surana,andK.Fall.Thecas

54、efortechnologyindevelopingregions.InIEEEComputer,June2005.6 A.Deshpande,C.Guestrin,S.Madden,andJ.Hellerstein.Model-drivendataacquisitioninsensornetworks.InProceedingsofVLDB,2004.7 A.Dhariwal,B.Zhang,B.Stau?er,C.Oberg,G.S.Sukhatme,D.A.Caron,andA.A.Requicha.Networkedaquaticmicrobialobservingsystem.InI

55、EEEIntl.Conf.onRoboticsandAutomation,May2006.8 ThomasC.Harmon.Microsensorstothemodelforecasts:Multiscaleembeddednetworkedsensingofnutrientsinthewatershed.EosTrans.AmericanGeophysicalUnion,FallMeet.Suppl.86(52):H23G-08(AbstractNo),2005.149 ThomasC.Harmon,RichardF.Ambrose,RobertM.Gilbert,JasonC.Fisher,MichaelSt