文章基本信息

标题：DETECTION OF UNDERDETERMINED SOURCE SEPARATION USING GASSIAN PROCESSES
本地全文：下载
作者：JAMRUD KHAN ; Dr.S.MADHAVA KUMAR ; Dr.S.BHARGAVI 等
期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2013
卷号：2
期号：5
页码：1732
出版社：S&S Publications
摘要：In this paper, we have stated the linear underdetermined, instantaneous, convolutive and multiple-outputsource separation problems in terms of Gaussian processes regression and. The advantages of setting out the sourceseparation problem in terms of GP are numerous. First, there is neither notational burden nor any conceptual issueraised when using input spaces X different from R or Z, thus enabling a vast range of source separation problems to behandled within the same framework. Multi-dimensional signal separation may include audio, image or video sensorarrays as well as geostatistics. Secondly, GP source separation can perfectly be used for the separation of non locallystationarysignals. Of course, some important simplifications of the computations as are lost when using non-stationarycovariance functions. Thirdly, it provides a coherent probabilistic way to take many sorts of relevant prior informationinto account.