kok电子竞技

1HowtoWritetheAbstractSection

孟帆159384吴文谦159387何恒涛159382AcademicWritingApril2220162ContentHowtostartElementsofanabstractLinguisticformatOthertipsExamplesHowtostart

Anabstractissimplyashort,standalonesummaryoftheworkorpaperthatotherscanuseasanoverview.

Anabstractdescribeswhatyoudoinyouressay.Itwillhelpyourreadertounderstandthepaperanditwillhelppeoplesearchforaparticularworktofinditanddecidewhetheritsuitstheirpurposes.3GettingYourAbstractStarted

Writeyourpaperfirst.

Anabstractworkstoreviewtheentiretyofthepaper,includingthemethodsandresults.itwillbeanoverviewofeverythingyouwriteaboutinyourpaper.Savewritingyourabstractforlast,afteryouhavealreadyfinishedyourpaper.Youwillbeabletogiveamuchmoreaccuratesummaryifyoudojustthat-summarizewhatyou'vealreadywritten.4Reviewandunderstandanyrequirementsforwritingyourabstract.

Thepaperyou’rewritingprobablyhasspecificguidelinesandrequirements,whetherit’sforpublicationinajournal,submissioninaclass,orpartofaworkproject.Beforeyoustartwriting,refertotherubricorguidelinesyouwerepresentedwithtoidentifyimportantissuestokeepinmind.5Consideryouraudience.

Forexample,inscientificjournals,abstractsallowreaderstoquicklydecidewhethertheresearchdiscussedisrelevanttotheirowninterest.Abstractalsohelpyourreadersgetatyourmainargumentquickly.Keeptheneedsofyourreadersinmindasyouwritetheabstract.

6ElementsofanAbstractDespitethefactthatanabstractisquitebrief,itmustdoalmostasmuchworkasthemultipaperthatfollowsit.Thismeansthatitshouldinmostcasesincludethefollowingsections.MotivationProblemMethodResultConclusion7

Motivation

Identifyyourpurpose.

Thereaderwantstoknowwhyyourresearchisimportant,andwhatthepurposeofitis.Startoffyourdescriptiveabstractbyconsideringthefollowingquestions:

Whydidyoudecidetodothisstudyorproject?Howdidyouconductyourresearch?Whatdidyoufind?Whyisthisresearchandyourfindingsimportant?8

Problem

Explaintheproblemathand.

Thinkofthisasthespecificissuethatyourresearchorprojectaddresses.Youcansometimescombinetheproblemwithyourmotivation,butitisbesttobeclearandseparatethetwo.

Whatproblemisyourresearchtryingtobetterunderstandorsolve?Whatisthescopeofyourstudy-ageneralproblem,orsomethingspecific?9

Method

Explainyourmethods.

Nowisthepartwhereyougiveanoverviewofhowyouaccomplishedyourstudy.Ifyoudidyourownwork,includeadescriptionofithere.Ifyoureviewedtheworkofothers,itcanbebrieflyexplained.

DescribetheevidenceyouhavetosupportyourclaimGiveanoverviewofyourmostimportantsources.10

Result

Describeyourresults(informativeabstractonly).

Inaninformativeabstract,youwillbeaskedtoprovidetheresultsofyourstudy.Whatisitthatyoufound?

Whatanswerdidyoureachfromyourresearchorstudy?Whatarethegeneralfindings?11

Conclusion

Giveyourconclusion.

Thisshouldfinishupyoursummaryandgiveclosuretoyourabstract.Init,addressthemeaningofyourfindingsaswellastheimportanceofyouroverallpaper.Thisformatofhavingaconclusioncanbeusedinbothdescriptiveandinformativeabstract,butyouwillonlyaddressthefollowingquestionsinaninformativeabstract.

Whataretheimplicationsofyourwork?Areyourresultsgeneralorveryspecific?121313TenseWeusuallyusethepresentsimpletenseandthepastsimpletensetogetaconciseabstract.Method

Usethepasttensetodescribethemethodsemployed. e.g.Weemployedconvexoptimizationtechniquestodevelop…Results

Presenttheminthepasttense. e.g.Wederivedaclosed-formtightupperboundforergodiccapacity.Conclusions

Usethepresenttensetostatetheprincipalconclusion. e.g.Itisshownthatwhenonlystatisticalinformationofchannel…Past

tense

is

used

to

describe

what

was

done

and

what

was

found!1414VoiceBeing

coherent

means

that

an

abstract

is

logical

and

make

sense.

A

good

abstract

uses

tense

correctly,

and

avoids

too

many

changes

between

active

voice

and

passive

voice.Thepast

PassivevoiceThepresent

Activevoice e.g.“This

paper

discusses

…”isbetterthan“Itis discussedbytheauthorinthepaperthat…” e.g.“Asetofequationswereproposed…”isbetterthan “Weproposedasetofequations…”1515SentencepatternTopicsentence Thetopicsentencealwaysindicatestheprimaryobjectiveofthe research.somesentencepatternsareasfollows:Thepurposeofthisresearchis…Inthispaper,weinvestigate…Thechiefaimofthepresentworkistoinvestigate…Theexperimentbeingmadebyourresearchgroupisaimedat…e.g.Weinvestigatemultiple-inputmultiple-output(MIMO)eigenmodetransmissionusingstatisticalchannelstateinformationatthetransmitter.

1616SentencepatternSupportingsentences Supportingsentencesfurtherspecifytheresearchtobepresented. Somethinglikebackground,methodandresultswillbeprovided inthispart.somesentencepatternsareasfollows:Although(theresearchsubject)…,(therelatedproblem)…isasyetundetermined.Severalsetsofexperimentshavebeenperformedtotestthevalidityof…Theresultsareasfollows…Theanalysisofthesamplesindicatesthat…e.g.Wealsoshowthesecrecysum-rateadvantageofoptimalpowerallocationoverequalpowerallocation.

1717SentencepatternConcludingsentences Astheendingpartofanabstract,concludingsentencesusually analyzetheresults,explaintheapplication,andpointoutthe significanceoftheresearch.somesentencepatternsareasfollows:Wepointsout/concludesthat…Theresultsareasfollows…Asaresultofthecurrentexperiment,weconcludethat…e.g.Motivatedbythisfact,wedesignanRCI-hybridschemebasedontheRCIandanewlyproposedhybridreceiverarchitecture.

Varietyofformsinvocabulary,expressions,structures,sentencespatternsarepreferablewhilesuchphraseshouldbeavoidedas“Ithink…”,”“Mystudy…”,“It`s…”!OtherTipsAnabstractmustbeafullyself-contained,capsuledescriptionofthepaper.Itcan'tassume(orattempttoprovoke)thereaderintoflippingthroughlookingforanexplanationofwhatismeantbysomevaguestatement.Itmustmakesenseallbyitself.Somepointstoconsiderinclude:18AvoidbeingtoospecificAnymajorrestrictionsorlimitationsontheresultsshouldbestated,ifonlybyusing"weasel-words"suchas"might","could","may",and"seem".19Usekeyphrasesandwords.

Ifyourabstractistobepublishedinajournal,youwantpeopletobeabletofinditeasily.Inordertodoso,readerswillsearchforcertainqueriesononlinedatabasesinhopesthatpapers,likeyours,willshowup.Trytouse5-10importantwordsorphraseskeytoyourresearchinyourabstract.

20Keepitinorder.

Therearespecificquestionsyourabstractmustprovideanswersfor,buttheanswersshouldbekeptinorderaswell.Ideally,itshouldmimictheoverallformatofyouressay,withageneral‘introduction’,‘body’and‘conclusion’.Manyjournalshavespecificstyleguidesforabstracts.Ifyou’vebeengivenasetofrulesorguidelines,followthemtotheletter.21Writeitfromscratch.

Yourabstractisasummary,yes,butitshouldbewrittencompletelyseparatefromyourpaper.Don'tcopyandpastedirectquotesfromyourself,andavoidsimplyparaphrasingyourownsentencesfromelsewhereinyourwriting.Writeyourabstractusingcompletelynewvocabularyandphrasestokeepitinterestingandredundancy-free.22Besuretodobasicrevisions.

Theabstractisapieceofwritingthat,likeanyother,shouldberevisedbeforebeingcompleted.Checkitoverforgrammaticalandspellingerrorsandmakesureitisformattedproperly.23Getfeedbackfromsomeone.

Trytofindsomeonewhodoesn’tknoweverythingaboutyourproject.Askhimorhertoreadyourabstractandthentellyouwhats/heunderstoodfromit.Thiswillletyouknowwhetheryou’veadequatelycommunicatedyourkeypointsinaclearmanner.24Example

Theestimationofsparseshallow-wateracousticcommunicationchannelsandtheimpactofestimationperformanceontheequalizationofphasecoherentcommunicationsignalsareinvestigated.Givensufficientlywidetransmissionbandwidth,theimpulseresponseoftheshallow-wateracousticchannelisoftensparseasthemultipatharrivalsbecomeresolvable.Inthepresenceofsigni?cantsurfacewaves,themultipatharrivalsassociatedwithsurfacescattering?uctuaterapidlyovertime,inthesensethatthecomplexgain,thearrivaltime,andtheDopplersofeacharrivalallchangedynamically.Purpose:?25Problem

Asparsechannelestimationtechniqueisdevelopedbasedonthedelay-Doppler-spreadfunctionrepresentationofthechannel.Thedelay-Doppler-spreadfunctionmaybeconsideredasa?rst-orderapproximationtotherapidlytime-varyingchannelinwhicheachchannelcomponentisassociatedwithDopplershiftsthatareassumedconstantoveranaveraginginterval.Thesparsestructureofthedelay-Doppler-spreadfunctionisthenexploitedbysequentiallychoosingthedominantcomponentsthatminimizealeastsquareserror.Theadvantageofthisapproachisthatitcapturesboththechannelstructureaswellasitsdynamicswithouttheneedofexplicitdynamicchannelmodeling.AsthesymbolsarepopulatedwiththesampleDopplers,theincreaseincomplexitydependsonthechannelDopplerspreadandcanbesigni?cantforaseverelyDoppler-spreadchannel.26Method

Comparisonismadebetweennonsparserecursiveleastsquares(RLS)channelestimation,sparsechannelimpulseresponseestimation,andestimationusingtheproposedapproach.Theresultsaredemonstratedusingexperimentaldata.Intrainingmode,theproposedapproachshowsa3-dBreductioninsignalpredictionerror.Indecision-directedmode,itimprovessigni?cantlytherobustnessoftheperformanceofthechannel-estimate-basedequalizeragainstrapidchannel?uctuations.IndexTerms—Channel-estimate-basedequalization,delayDoppler-spreadfunction,matchingpursuit,sparseestimation,time-varyingchannels.27Conclusion（presenttense）Result