权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Signal separation from the mixture of correlated multiple signals and its application

相关多信号混合信号的分离及其应用

基本信息

批准号：
12450163
负责人：
KOBATAKE Hidefumi
金额：
$ 5.44万
依托单位：
TOKYO UNIVERSITY OF AGRICULTURE AND TECHNOLOGY
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (B)
财政年份：
2000
资助国家：
日本
起止时间：
2000 至 2001
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-12450163/
关键词：
BSS Signal Separation Independent Component Corre l at I on 音源同定近接音場 MUSIC法広帯域信号

项目摘要

The important results of this research projects can be summarized as follows.Energy sources localization problem has not been fully addressed in nearfield scenario. We have presented an eigen-structure based near-field wideband sources localization method using arbitrarily spaced sensor array. Estimation of the source locations is based on the straightforward exploitation of the eigenstructure of array power spectral density matrices. Our method solves the 3-D (azimuth, elevation and range) localization problem. A fast algorithm but with a sacrifice in the freedom in sensor arrangement is also presented. Simulation tests prove its validity and good performance.Blind separation of convolutive mixtures is often implemented in rfequency domain. The partial correlation between the frequency representations of original sources is sometimes too high to be neglected and consequently makes the performance of all the BSS methods efficiency down in various degrees. We proposed a recursive blind separation method for lowering the unfavorable effect from the partial correlation. The signal-to-noise ratio is greatly increased at certain bins, which results in a much better separation. Theoretically, it is possible to be used combined with any blind separation method for which the independence is assumed, and better separation can be expected.

本研究的重要成果可以概括如下：近场情况下的能源定位问题尚未得到充分解决。提出了一种基于特征结构的任意间距传感器阵列近场宽带源定位方法。源位置的估计是基于阵列功率谱密度矩阵的特征结构的直接开发。我们的方法解决了3-D（方位角，仰角和范围）定位问题。文中还提出了一种牺牲传感器布置自由度的快速算法。仿真实验证明了该方法的有效性和良好的性能。卷积混合信号的盲分离通常在频域实现。原始信号频率表示之间的部分相关性有时太高而不能忽略，从而使所有BSS方法的性能都不同程度地下降。为了降低部分相关对分离的不利影响，提出了一种递归盲分离方法。信噪比在某些频段大大增加，这导致更好的分离。从理论上讲，它可以与任何假设独立性的盲分离方法结合使用，并且可以期望更好的分离。