Logo image
Two-Way Training Design for Discriminatory Channel Estimation in Wireless MIMO Systems
Thesis

Two-Way Training Design for Discriminatory Channel Estimation in Wireless MIMO Systems

Huang, Chao-Wei
Masters, 國立清華大學, 通訊工程研究所
2010

Abstract

人工雜訊 通道估計 artificial noise channel estimation
Thisworkexaminestheuseoftwo-waytraininginmultiple-inputmultiple-output(MIMO)wirelesssystemstodiscriminatethechannelestimationperformancesbetweenalegitimatereceiver(LR)andanunauthorizedreceiver(UR).Thisthesisextendsuponthepreviouslyproposeddiscriminatorychannelestimation(DCE)schemethatallowsonlythetransmittertosendtrainingsignals.ThegoalofDCEistominimizethechannelestimationerroratLRwhilerequiringthechannelestimationerroratURtoremainbeyondacertainlevel.Ifthetrainingsignalissentonlybythetransmitter,theperformancediscriminationbetweenLRandURwillbelimitedsincethetrainingsignalshelpbothreceiversperformestimatesoftheirdownlinkchannels.Inthiswork,weconsiderinsteadthetwo-waytrainingmethod-ologythatallowsboththetransmitterandLRtosendtrainingsignals.Inthiscase,thetrainingsignalsentbyLRhelpsthetransmitterobtainknowledgeofthetransmitter-to-LRchannel,butdoesnothelpURestimateitsdownlinkchannel(i.e.,thetransmitter-to-URchannel).Withtransmitterknowledgeoftheestimatedtransmitter-to-LRchannel,artificialnoise(AN)canthenbeembeddedinthenullspaceofthetransmitter-to-LRchanneltodisruptUR’schannelestimationwithoutseverelydegradingthechannelestimationatLR.Basedontheseideas,two-wayDCEtrainingschemesaredevelopedforbothreciprocalandnon-reciprocalchannels.TheoptimalpowerallocationbetweentrainingandANsignalsisdevisedunderbothaverageandindividualpowerconstraints.Numericalresultsareprovidedtodemonstratetheefficacyoftheproposedtwo-wayDCEtrainingschemes.

Metrics

1 Record Views

Details

Logo image