{"title":"On detection with a class of matched filters and higher-order statistics","authors":"B. Sadler, G. Giannakis","doi":"10.1109/SPECT.1990.205579","DOIUrl":null,"url":null,"abstract":"The authors of an earlier paper (G.B. Giannakis and M.K. Tsatsanis, IEEE Trans. Acoust. Speech Signal Process, vol.38, no.7, p.1284-1990). Propose examining the zero-th lag of the higher-order (>2) cumulant of the output of a matched filter for detection. This approach exploits the desirable properties of cumulants including their insensitivity to additive colored Gaussian noise of unknown covariance. A binary hypothesis test is formulated which chooses between two Gaussian distributions which arise from the asymptotic cumulant estimator's behavior. The detection results of the earlier paper are extended to a class of filters which include the matched filter, phase-only matched filters, and inverse filters.<<ETX>>","PeriodicalId":117661,"journal":{"name":"Fifth ASSP Workshop on Spectrum Estimation and Modeling","volume":"55 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Fifth ASSP Workshop on Spectrum Estimation and Modeling","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SPECT.1990.205579","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
The authors of an earlier paper (G.B. Giannakis and M.K. Tsatsanis, IEEE Trans. Acoust. Speech Signal Process, vol.38, no.7, p.1284-1990). Propose examining the zero-th lag of the higher-order (>2) cumulant of the output of a matched filter for detection. This approach exploits the desirable properties of cumulants including their insensitivity to additive colored Gaussian noise of unknown covariance. A binary hypothesis test is formulated which chooses between two Gaussian distributions which arise from the asymptotic cumulant estimator's behavior. The detection results of the earlier paper are extended to a class of filters which include the matched filter, phase-only matched filters, and inverse filters.<>