基于RSAMP算法的OFDM稀疏信道估计

为了提高稀疏度自适应贪婪迭代(sparsity adaptive greedy iterative, SAGI)算法的重构性能, 缩短重构时间, 提出了一种基于有限等距性质(restricted isometry property, RIP)的稀疏度预测自适应匹配追踪(RIP based prediction-sparsity adaptive matching pursuit, RSAMP)算法, 并成功将其应用于正交频分复用(orthogonal frequency division multiplexing, OFDM)系统信道估计。首先, 提出一种基于RIP的稀疏度预测方法, 可以在稀疏度未知的情况下快速精确地逼近真实稀疏度, 大大缩短了算法的运行时间。其次, 利用主成分分析法对观测矩阵采取了优化处理, 提高了算法的重构性能。仿真实验显示, 相较于SAMP、SAGI算法, 本文提出的RSAMP算法可以获取更好的估计性能和更短的运行时间。

OFDM sparse channel estimation based on RSAMP algorithm

In order to improve the reconstruction performance of the sparsity adaptive greedy iteration (SAGI) algorithm and shorten the reconstruction time, a restricted isometry property (RIP) based prediction-sparsity adaptive matching pursuit (RSAMP) algorithm is proposed and successfully applied to channel estimation of orthogonal frequency division multiplexing (OFDM) system. Firstly, a RIP based sparsity prediction method is proposed, which can approximate the real sparsity quickly and accurately in the case of unknown sparsity, greatly reducing the running time of the algorithm. Secondly, the observation matrix is optimized by principal component analysis, which improves the reconstruction performance of the algorithm. Simulation experiments show that the proposed RSAMP algorithm in this paper can achieve better channel estimation performance and shorter running time compared with SAMP algorithm and SAGI algorithm.