简介：Withtheemergenceoflocation-basedapplicationsinvariousfields,thehigheraccuracyofpositioningisdemanded.Byutilizingthetimedifferencesofarrival(TDOAs)andgainratiosofarrival(GROAs),anefficientalgorithmforestimatingthepositionisproposed,whichexploitstheBroyden-Fletcher-Goldfarb-Shanno(BFGS)quasi-Newtonmethodtosolvenonlinearequationsatthesourcelocationundertheadditivemeasurementerror.Althoughtheaccuracyoftwo-stepweighted-least-square(WLS)methodbasedonTDOAsandGROAsisveryhigh,thismethodhasahighcomputationalcomplexity.Whiletheproposedapproachcanachievethesameaccuracyandbiaswiththelowercomputationalcomplexitywhenthesignal-to-noiseratio(SNR)ishigh,especiallyitcanachievebetteraccuracyandsmallerbiasatalowerSNR.Theproposedalgorithmcanbeappliedtotheactualenvironmentduetoitsreal-timepropertyandgoodrobustperformance.Simulationresultsshowthatwithagoodinitialguesstobeginwith,theproposedestimatorconvergestothetruesolutionandachievestheCramer-Raolowerbound(CRLB)accuracyforbothnear-fieldandfar-fieldsources.