摘要： 正交频分复用系统中的信号在传输过程中受到无线信道环境衰落和延时的影响，容易产生符号间干扰（ISI），对信道状态信息进行准确估计是降低ISI、提高信号传输准确率的有效方法。针对贪婪迭代类压缩感知信道估计算法存在的估计径错误及漏选问题，提出一种基于离散傅里叶变换（DFT）寻径的压缩感知信道估计算法DFT-OMP。通过DFT寻径的方式抑制由噪声引起的不理想原子，从而对OMP算法重构过程中的原子进行筛选，解决传统方法选取相关因子最大的原子作为重构原子而导致的依赖信号稀疏度问题。在原子预选后的贪婪迭代类压缩感知算法信道估计中引入残差精度控制，以提高信道估计的自适应性与鲁棒性。仿真结果表明，相对OMP算法，该算法能取得4 dB的信道估计性能增益，其适用于较大导频下的无线通信系统。

关键词: 压缩感知, 信道估计, 贪婪迭代, 离散傅里叶变换寻径, 残差精度控制

Abstract: Signals in Orthogonal Frequency Division Multiplexing(OFDM) systems are influenced by environmental decline in wireless channels and delay, causing Inter Symbol Interference(ISI).Accurate estimation of Channel State Information(CSI) is an effective approach of reducing ISI and improving signal transmission accuracy.To address the estimation path errors and missing selection of the channel estimation algorithms based on greedy iterative compressed sensing, an improved compressed sensing channel estimation algorithm named DFT-OMP is proposed based on Discrete Fourier Transform(DFT) path searching.The algorithm employs DFT path searching to suppress the unpleasant atoms caused by noise, so the atoms generated by the reconstruction process in the OMP algorithm are filtered.Consequently, the problems of traditional compressed sensing channel estimation methods, which are caused by selecting the atom with the largest correlation factor, are solved, including sparsity of dependent signals.After atom preselection, the residual precision control method is introduced into the channel estimation part of the greedy iterative compressed sensing algorithm to improve the adaptability and robustness of channel estimation.The simulation results show that compared with the OMP algorithm, the proposed algorithm displays a 4 dB gain in channel estimation performance, and is suitable for wireless communication systems with large pilots.

Key words: Compressed Sensing(CS), channel estimation, greedy iteration, Discrete Fourier Transfrom(DFT) path searching, residual precision control

中图分类号: 

• TN911