基于子空间拟合的 UWB 稀疏信道盲估计

针对 TH-PPM 超宽带系统,利用信道的稀疏性,提出一种基于子空间拟合的稀疏信道盲估计算法.该算法首先利用接收信号的均值循环卷积特性,估计出信道的频域响应,接着利用子空间拟合算法确定出非零抽头系数的位置,最后再采用最小二乘算法完成对非零系数值的估计,避免了无谓的零抽头估计,改善了算法性能.该算法以码片周期采样,且仅需要作一些简单的信号处理如重叠相加、离散傅立叶变换等,因此运算量较小.仿真表明,与没有利用信道稀疏性的一阶盲算法相比具有更优的均方误差和误比特率性能.

Blind sparse channel estimation for Ultra Wide-Band communications based on representative subspace fitting

In this paper,a novel blind channel estimation algorithm is proposed for TH-PPM Ultra Wide-Band communications based on first-order statistics and representative subspace fitting by exploiting the sparsity property of channel impulse response.Firstly,the channel frequency response is derived by exploiting the cyclic convolution property of the receive signal's expectation.Secondly,the representative subspace fitting algorithm is developed to detect the position of the nonzero tap coefficients,and then the least square algorithm is used to estimate the exact value of the nonzero coefficients,so as to improve the performance of the algorithm by avoiding estimating the zero coefficients.The algorithm's complexity is low for using chip-rate sampling and only some overlap-add operations and discrete Fourier transform operations and so on are needed.Simulation results demonstrate that it has better MSE and BER performance than the first-order blind channel estimation which didn't make use of the sparse structure of the channel.