{"title":"Statistical Estimation of Uncertainty in Surface Duct Parameters Inversion","authors":"Hai-tian Zhao, Zhensen Wu","doi":"10.1109/ISAPE.2018.8634037","DOIUrl":null,"url":null,"abstract":"The tropospheric duct parameter inversion problem using the global optimization algorithm to find the optimal value of the objective function is a typical point estimation problem. In some cases, we need not only the best estimate of the refractive profile, but also the probability problem of the most advantageous, that is, the uncertainty of the inversion. Therefore, this paper uses Bayesian theory to calculate the uncertainty of the statistics of duct profile parameters. However, since the parameter vector dimension is high, the posterior probability density of each parameter cannot be directly calculated. Therefore, an efficient sampling algorithm is needed to sample the parameter vector. We used the Metropolis-Hasting sample in the MCMC sampling algorithm to sample the profile parameters and obtained a good statistical result.","PeriodicalId":297368,"journal":{"name":"2018 12th International Symposium on Antennas, Propagation and EM Theory (ISAPE)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 12th International Symposium on Antennas, Propagation and EM Theory (ISAPE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISAPE.2018.8634037","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
The tropospheric duct parameter inversion problem using the global optimization algorithm to find the optimal value of the objective function is a typical point estimation problem. In some cases, we need not only the best estimate of the refractive profile, but also the probability problem of the most advantageous, that is, the uncertainty of the inversion. Therefore, this paper uses Bayesian theory to calculate the uncertainty of the statistics of duct profile parameters. However, since the parameter vector dimension is high, the posterior probability density of each parameter cannot be directly calculated. Therefore, an efficient sampling algorithm is needed to sample the parameter vector. We used the Metropolis-Hasting sample in the MCMC sampling algorithm to sample the profile parameters and obtained a good statistical result.