针对短波单载波频域均衡( SC-FDE)系统中最小二乘( LS)信道估计算法受噪声影响大而导 致的估计精度低的问题,提出了 一种改进的基于小波去噪的LS信道估计算法.改进算法采用基于 分块导频的帧结构, 首先用LS算法对信道进行初步估计, 然后根据小波多分辨率分析( Mallat)理论 将 LS估计的结果分解, 并设置一个合理阈值对分解得到的小波系数处理, 从而消除LS估计的残留 噪声, 提高估计精度.仿真结果表明, 在短波信道下, 改进算法不仅减小了系统开销, 而且提高了 LS 估计的性能.
The Least Square ( LS) channel estimation algorithm for single carrier frequency domain equali-zation (SC - FDE) system in high frequency( HF) communication is affected by the noise,which leads to a lower estimation precision. To solve this problem,an improved LS channel estimation algorithm based on wavelet denoising is proposed. The new method uses the frame structure based on block pilot,decomposes the results of preliminary LS channel estimation according to wavelet multi-resolution analysis( Mallat) the-ory firstly,and sets a reasonable threshold to deal with the decomposed wavelet coefficients,thus eliminating the residual noise of LS channel estimation and improving the estimation precision. Simulation results in shortwave channel show that the proposed algorithm not only reduces the system's overhead,but also has a significant performance improvement compared with traditional LS channel estimation.