中文摘要：

在城市密集环境情况下，基于传统点目标模型的多径空时参数（波达方向与相对时延）联合估计算法往往出现性能恶化。本文通过重新构造空间流形与时间流形，给出了一种更符合密集环境的空时相干分布信号模型，并在获得空时联合信号子空间后，分别抽取空间、时间子矩阵构造空间谱与时间谱，通过搜索谱峰得到所需空、时多径参数，最后利用空时流形向量与空时联合噪声子空间的正交性质实现空时参数配对。与TST-MUSIC等算法相比，该算法仅需要两次搜索及一次配对过程。仿真实验表明，该方法能够有效估计相干分布信号空时参数，抑制分布扩展影响，实现多个参数的自动配对。

英文摘要：

Under dense urban fading environment, performance of joint multi-path parameter estimation method based on traditional point signal model degrades seriously. In this paper, after reconstructing space and time manifold, a new space and time signal model based on muhipath coherent distribution function is given. Space and time spectrum is constructed after space sub-matrix and time sub-matrix is taken out of acquired joint space-time signal subspace,then space and time parameters is find by searching space spectrum and time spectrum. In the end space-time parameter pairs is matched using orthogonal property between joint space-time noise subspace and space-time manifold vector. In contrast with TST-MUSIC, the algorithm we present here only need two searching and one pairing processes. Simulation results indicated that the proposed method can estimate space and time parameters, restrain space-time distribution and multi-parameters can be paired automatically.