| 基于共变谱和旋转不变技术的多径时延估计 |
| |
| 引用本文： | 刘文红,迟冬祥,李渊,孟银阔,钟旭. 基于共变谱和旋转不变技术的多径时延估计[J]. 上海电机学院学报, 2011, 14(4): 211-215 |
| |
| 作者姓名： | 刘文红 迟冬祥 李渊 孟银阔 钟旭 |
| |
| 作者单位： | 上海电机学院电子信息学院,上海,200240 |
| |
| 基金项目： | 中国博士后科学基金资助，上海电机学院重点培育项目资助，上海电机学院科研启动经费项目资助，上海电机学院重点学科资助 |
| |
| 摘 要： | 针对脉冲噪声环境下多径时延估计的分辨率问题,提出了一种基于共变谱(CS)和信号估计参数旋转不变技术(ESPRIT)的多径时延估计方法。采用α稳定分布建模脉冲噪声,依据分数低阶统计量理论,将两接收信号的共变序列进行傅里叶变换得到共变谱。CS看作等效的时间序列,把时域的多径时延估计问题转化为频域的多个正弦信号频率估计问题。利用ESPRIT对等效时间序列进行频率估计,得到高分辨率的多径时延估计。仿真实验表明:多径时延估计方法(CS_ESPRIT)在高斯和脉冲噪声环境中均具有较好的估计性能。
|
| 关 键 词： | α稳定分布 共变谱 信号估计参数旋转不变技术 多径时间延迟 |
Multipath Time Delay Estimation Based on CS_ESPRIT |
| |
| LIU Wenhong,CHI Dongxiang,LI Yuan,MENG Yinkuo,ZHONG Xu. Multipath Time Delay Estimation Based on CS_ESPRIT[J]. JOurnal of Shanghai Dianji University, 2011, 14(4): 211-215 |
| |
| Authors: | LIU Wenhong CHI Dongxiang LI Yuan MENG Yinkuo ZHONG Xu |
| |
| Affiliation: | LIU Wenhong,CHI Dongxiang,LI Yuan,MENG Yinkuo,ZHONG Xu(School of Electronic and Information,Shanghai Dianji University,Shanghai 200240,China) |
| |
| Abstract: | To deal with resolution of multipath time delay estimation in the impulsive noise, a method for estimating multipath time delay is proposed based on covarianee spectrum （CS） and estimation signal parameter via rotational invariance technique（ESPRIT）. The α-stable distribution is used to model the impulsive noise. CS is obtained from FFT of the covariance sequence of the two received signals based on the theory of fractional lower order statistics. Thus, multipath time delay estimation is converted to a problem of multi-frequency estimation for CS is a sequence in the time domain. High resolution multipath time delay can be obtained with ESPRIT. The method is termed CP-ESPRIT in this paper. Computer simulation shows that the method has satisfactory performance both in Gaussian and impulsive noise environments. |
| |
| Keywords: | α-stable distribution covariance spectrum（CS） estimation signal parameter viarotational invariance technique（ESPRIT） multipath time delay estimation |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
