浅海地声参数的变维贝叶斯反演*吴伟文1,2任群言1,2†鹿力成1,2郭圣明1,2马力1,2(1中国科学院水声环境特性重点实验室(声学研究所)北京100190)(2中国科学院大学北京100049)2022年6月6日收到2022年11月2日定稿摘要为解决海底沉积层分层结构未知时的反演问题,提出了一种变维粒子滤波方法,利用声场的互谱密度,估计沉积层分层结构以及地声参数。仿真结果表明：变维粒子滤波在沉积层分层结构未知时,能有效反演沉积层层数以及地声参数,粒子的并行计算能使其相较于可逆跳蒙特卡洛马尔可夫链(rjMCMC)更加高效。利用变维粒子滤波,对南海垂直线阵列接收到的线性调频信号进行处理,反演结果与rjMCMC反演得到的沉积层层数和地声参数结果相近,说明了此方法能有效估计沉积层层数的同时反演浅海地声参数,得到可靠的参数后验概率密度。关键词地声反演,地声模型,贝叶斯理论,不确定性分析PACS数43.30,43.60Trans-dimensionalBayesiangeoacousticinversioninshallowwaterWUWeiwen1,2RENQunyan1,2†LULicheng1,2GUOShengming1,2MALi1,2(1KeyLaboratoryofUnderwaterAcousticEnvironment,InstituteofAcoustics,ChineseAcademyofSciencesBeijing100190)(2UniversityofChineseAcademyofSciencesBeijing100049)ReceivedJun.6,2022RevisedNov.2,2022AbstractTodealwiththeinversionproblemwhenthespatialstructureofseafloorsedimentsisunknown,atrans-dimensionalparticlefiltermethodisproposedinthispaper,wherethecross-spectraldensityofthepressurefieldisusedtoestimatethesedimentlayeringstructureandgeoacousticparameters.Thesimulationresultsshowthatthenumberofsedimentlayersandgeoacousticparameterscanbeeffectivelyestimatedusingtheproposedmethod,andtheparallelparticlecalculationmakethismethodmoreefficientthanreversiblejumpMonteCarloMarkovchain(rjMCMC)sampling.Alinearfrequency-modulatedsignalreceivedbyaverticallinearrayintheSouthChinaSeaisprocessedusingtheproposedtrans-dimensionalparticlefilter.TheinversionresultsofthesedimentlayersandgeoacousticparametersaresimilartothoseacquiredbyrjMCMCinversion.Thenumberofsedimentlayersandtheposteriorprobabilitydensityofparameterscanbeeffectivelyestimatedusingthismethod.KeywordsGeoacousticinversion,Geoacousticmodel,Bayesiantheory,Uncertaintyanalysis引言声信号在水中具...
