Low-complexity method for DOA estimation based on ESPRIT

在线阅读 下载PDF 导出详情
摘要 Alow-complexitymethodfordirectionofarrival(DOA)estimationbasedonestimationsignalparametersviarotationalinvariancetechnique(ESPRIT)isproposed.Insteadofusingthecross-correlationvectorsinmultistageWienerfilter(MSWF),theorthogonalresidualvectorsobtainedinconjugategradient(CG)methodspanthesignalsubspaceusedbyESPRIT.Thecomputationalcomplexityoftheproposedmethodissignificantlyreduced,sincethesignalsubspaceestimationmainlyneedstwomatrixvectorcomplexmultiplicationsattheiterationofdatalevel.Furthermore,thepriortrainingdataarenotneededintheproposedmethod.Toovercomeperformancedegradationatlowsignaltonoiseratio(SNR),theexpandedsignalsubspacespannedbymorebasisvectorsisusedandsimultaneouslyrendersESPRITyieldredundantDOAs,whichcanbeexcludedbyperformingESPRIToncemoreusingtheunexpandedsignalsubspace.ComparedwiththetraditionalESPRITmethodsbyMSWFandeigenvaluedecomposition(EVD),numericalresultsdemonstratethesatisfactoryperformanceoftheproposedmethod.
机构地区 不详
出版日期 2010年05月15日(中国期刊网平台首次上网日期,不代表论文的发表时间)
  • 相关文献