正交频分复用系统中基于子空间跟踪的信道估计

基于导频的正交频分复用（OFDM）系统信道估计方法是通过在确定的时频位置发送已知的导频符号来获取信道参数的，通常包括导频位置的最小二乘估计和内插滤波两步。如果能够提高导频位置信道频域特性的估计精度，那么后续内插滤波的精度也就能相应地得到提高，为此，在建立通用参数化信道模型的基础上，提出一种子空间跟踪的改进算法，该算法通过基于Givens旋转的时延子空间跟踪和基于RLS滤波的幅度跟踪，提高了导频符号信道估计的精度，降低了跟踪算法的复杂度。仿真结果表明，基于子空间跟踪的信道估计方法具有较低的复杂度和较高的估计精度。

A Channel Estimation Method for Orthogonal Frequency Division Multiplexing Systems by Tracking Delay-Subspace

Channel parameters in orthogonal frequency division multiplexing (OFDM) systems over fading channels are generally obtained by transmitting pilot symbols in given positions of the time-frequency grid. The pilot-based method usually involves the LS estimation step on the pilot subcarriers and the interpolation step over the entire time-frequency grid. Increasing the accuracy of the pilot estimation will accordingly increase the precision in the interpolation step. This paper proposed an improved subspace-tracking algorithm by the Givens rotation based delay-subspace tracking and the RLS filtering based amplitudes tracking. Performed on the parameterized model of the channel, the simulations prove that the channel (estimation) is more robust and accurate while the complexity of the algorithm is low.