SUetal.:ACTIVECOOPERATIONBETWEENPRIMARYUSERSANDCRUSERScouldembedandtransmittheirowndatawhileforwardingun-successfultransmittedpacketsofprimaryuser.In[24],aspec-trumleasingschemewasproposedinwhichprimaryusersmayleasewholeownedbandwidthforafractionoftimetoanadhocnetworkofsecondaryusersbasedondecode-and-forwardfor-wardingschemeanddistributedspace-timecoding.In[25],itwasproposedtodeploya“dumb”relaynodeincognitiveradionetworkstohelprelayprimaryorCRusersignalstoimprovenetworkspectrumef ciency,wheretheprotocolwasanalyzedandoptimizedforanetworkmodelconsistingofapairofpri-maryusersandapairofCRusers.
Inthispaper,weconsideractivecooperationbetweenpri-maryusersandCRusersinheterogeneousnetworks.Incontrastwithexistingliterature[20],wemakearealisticassumptionthatprimaryuserdataareneitherknownnorperfectlydecodedbyCRusers.WeproposeanactivecooperationschemebetweenprimaryusersandCRuserswhichallowsformorerobust(po-tentiallycontinuous)service/operationoftheCRusersinthenetwork.Thisis,again,incontrastwithexistingwork[22]–[25]whereCRuserstransmittheirownsignalsonlyifprimaryusersareidle,resultinginCRdatatransmissionthatmaybeinter-ruptedfrequently.Intheproposedprotocol,primaryusersandCRuserscooperateformutualbene t—CRusersassisttorelayprimaryusersignalsinexchangeforsomespectrumreleasedbytheprimaryusers.Weunderstandthatsuccessfuldeliveryofdataatthelowerpossiblepowercostis,arguably,themostim-portantobjectiveofaprimaryuserwhilethepercentageofband-widthusedfordatatransmissionisnotcritical.Asaresult,theprimaryusermayreleaseaportionofthebandwidthtotheCRuserfortransmissionofitsowndata.TheamountofthereleasedbandwidthmaydirectlydependontheCRusereffortneededtorelaytheprimaryusersignals.Immediatebene tsofferedtotheprimaryuserbytheabovecooperationprotocolincludetrans-missionpowersavingsandon-airtimereductionswhichcanbecritical,forexample,inlow-probability-interception(LPI)andlow-probability-detection(LPD)militaryapplications.
Morespeci cally,inthispaperwedesignacognitivecoop-erationprotocolthatmaximizestheprimaryuserpowersavingsandtheCRuserowndatatransmissionrate.First,wedevelopanoptimumpowerallocationschemefortheCRuserthatmax-imizestheCRuser’sowndatatransmissionrateforanygivenpowerbudgetandagivenbandwidthreleasedfromtheprimaryuser.Thenwedeterminearangeofpossibletransmissionpowerlevelsfortheprimaryusertooperatewithoutsacri cingitsownbene tincooperation.Wealso ndanecessaryconditiononthechannelqualitythatallowscooperationbetweentheprimaryuserandtheCRuser.Asitturnsout,cooperationistriggeredonlyiftheamplitudeofthechannelbetweentheprimaryuserandtheCRuserisaboveacertainthreshold.Numericalresultsshowthattheprimaryuserenjoyssigni cantaveragepowersavingsasaresultofthecooperationwithCRusers.Itmaysavepowerupto80%comparedtothenoncooperationcase(theexactpercentageofpowersavingsdependsonthechannelcon-dition).
Therestofthispaperisorganizedasfollows.InSectionII,wespecifyafour-nodecognitiveradionetworkanddescribetheproposedcognitivecooperationprotocol.InSectionIII,weoptimizetheCRuserrelayingpowertomaximizeitsowndatatransmissionrate,basedonitspowerbudget.InSectionIV,wediscussprimaryuseroptionsforcooperation,anddetermine

a
1797
Fig.1.Time-bandwidthallocationforPUandCRusers.
minimumnecessarypowerlevelaswellasamaximumpossiblepowerlevelfortheprimaryusertoinitiatecooperationwithoutsacri cingitsowntransmissionrequirement.InSectionV,wecomparevariousstrategiesfortheprimaryuserpoweralloca-tionintermsofprimaryuserpowersavingsandCRuserowndatatransmissionrate.Finally,someconclusionsaredrawninSectionVI.
II.PROPOSEDCOGNITIVECOOPERATIONPROTOCOL
Forsimplicityinpresentation,weconsiderabasicfour-nodecognitiveradionetworkconsistingofapairofprimaryusersandapairofCRusers1.WeassumethatthebandwidthownedbytheprimaryusersisHz.Foreachprimaryuser,wealsoassume
thatthetargettransmissionrateisbits/sandthetransmis-sionpowerperunitfrequencyisWatts/Hz.IfCRusersareavailabletoassistwithrelayingprimaryuserdata,primaryusersmayreleasesomespectrumtoCRusersfortheirownuse,asshowninFig.1.Speci cally,letdenotethetimedurationthataprimaryuserisallowedtotransmitdataoverbandwidthHz.IfaCRuserisavailabletoassist,thentheprimaryuser,withoutcompromisingitstargettransmissionrate,maydecidetotransmitinonlyapartofthetimeslot(e.g.,)overaportion
ofthebandwidth,
(Phase1),hopingcooperationwithaCRuseranddatarelayingfortherestofthetimeslot(Phase2)mayresultinsuccessfuldeliveryofprimaryuserdatatotheintendeddestination.Theimmediatebene tsofcooper-ationfortheprimaryuserateon-airtimereductionandtrans-missionpowersavings.Atthesametime,theCRuserassiststorelayprimaryusersignalsinexchangeforsomebandwidth,Hz,releasedbytheprimaryuser.
LetusassumethattheprimaryuserutilizesbandwidthHz
andreleasesbandwidth
HztoCRusers.Letdenotethesignaltransmittedbytheprimaryuser(source).Thetransmittedsignalmaybereceivedbyboththeintendedprimaryuser(destination)andanearbyCRuser(relay).Thereceivedsignalsatthedestinationandattherelayaregivenby
(1)(2)
where
anddenotethechannelsbetweentheprimarysourceandtheandbetweenthesourceandtheCR
1The
proposedcognitivecooperationprotocolandthetheoreticaldevelop-mentinthispapercanbegeneralizedtocognitiveradionetworkswithmoreprimaryusersandmoreCRusers,inwhicheachprimaryusermaychoosesev-eralCRusersorthebestCRuserforcooperation.
Optimal spectrum allocation of primary users in light-handed cognitive radio networks_专业资料。Jnl厂ab stefTho (eee)—11,—321 or Hrnnito cnlaDiItu ...
Primary research on ne... 1人阅读 3页 2.00 Effect of sample diame... 2人阅读 8页 2.00 Bifurcation control of... 1人阅读 6页 2....
“talk under” primary users in a given ...Users utilizing Cognitive Radio technology This ...Foster Cooperation Expand Unlicensed Spectrum Relax ...
Scheduling in Cooperate-to-Join Cognitive Radio ...slot with primary nodes in return for cooperation... a cognitive node half way between the licensed ...
active monitoring of several factors in the ...for exchanging spectrum sensing data between nodes....Cognitive radio users are incorporated for primary ...
A Survey of Cognitive Radio Technologies and_...active primary user in a licensed band (i.e.,... to avoid collision between unlicensed users. ...
between cognitive users through competing and ...96 5.3 Asymmetric active cooperation strategy in ...primary-user problem 与集中式网络相比,在分布式网络...
about primary users’ parameters available at cognitive users are considered. ...the Kullback-Leibler distance between signal plus noise and noise only models...
cooperation between a PU link and a SU link in... if PU is active, SU try to receive the data...Cognitive Radio with Multi-Access Primary ...
COGNITIVE RADIO NETWORKS Sachin Shetty, Kodzo ...cooperation with the objective of minimizing ...The number of primary users in the network is ...
热门文档