一种新的码片内多径时延估计方法

摘要

根据信道冲激响应的稀疏特性,提出了一种频域的时延估计压缩感知模型,将时延估计问题转化为基于欠采样数据的稀疏向量估计问题.利用离散傅里叶变换(Discrete Fourier transform,DFT)矩阵的子矩阵所满足的受限等距性(Restricted isometry property,RIP)以及信道冲激响应的稀疏特性充分降低了时延估计所需数据量的要求.分析了本文模型具有码片内多径分辨能力以及良好抗噪性能的原因,并与多信号分类(Multiple signal classification,MUSIC)和旋转不变技术的信号参数估计(Estimation of signal parameters via rotational invariance technique,ESPRIT)算法的时延估计性能进行仿真比较.仿真结果表明,本文提出的方法不需要预知多径的条数,对码片内多径时延具有较高的估计精度,其时延估计性能在特定条件下优于MUSIC和ESPRIT算法.%According to the sparsity of the channel response, a multipath time-delay estimation model based on compressed sensing (CS) is proposed in frequency domain. Using the model, the time-delay estimation is converted into sparse vector estimation from undersampled data. The restricted isometry property (RIP) is satisfied by partial discrete Fourier transform (DFT) matrix and the channel response is sparse. According to CS theory, the required amount of data for the estimation is sharply decreased. The reason why the proposed method has subchip multipath estimation ability and excellent anti-noise property is also analyzed. Then, the time-delay estimation performance of the CS method is compared with the multiple signal classification (MUSIC) algorithm and the estimation of signal parameters via rotational invariance technique (ESPRIT) algorithm. Simulation and analysis results show that the pro posed method has super-resolution performance in subchip multipath time-delay estimation ow ing to no prediction of the number of multipath. And it is superior to MUSIC and ESPRIT al gorithms under certain conditions.