MIMO-OFDM系统中时变稀疏信道估计  

Time-Varying Sparse Channel Estimations in MIMO-OFDM Systems

作　　者：荆楠[1] 王林[1] 

机构地区：[1]燕山大学信息科学与工程学院,河北秦皇岛066004

出　　处：《信号处理》2016年第8期889-897,共9页Journal of Signal Processing

基　　金：国家自然科学基金(61303233);河北省高等学校科学技术研究青年基金(QN20131058);河北省自然科学基金(F2016203176)

摘　　要：在实际移动通信系统中,信道多径的数量、多径的幅值及多径时延均会随着移动台的运动而动态发生变化,传统的稀疏信道估计方法将不再适用。针对MIMO OFDM系统的时变稀疏信道模型,该文提出一种基于卡尔曼滤波的时变稀疏信道估计算法。该算法将重加权最小1范数问题转化为非线性等式约束卡尔曼滤波问题,并构造一个伪观测方程线性化重加权最小1范数约束方程。然后,分两次采用卡尔曼滤波器实现时变稀疏信道的重构。仿真结果表明,在信道稀疏度动态变化的情况下,本文算法的信道估计均方误差优于基追踪等经典的稀疏信道估计算法。In the practical mobile communication systems, the number of muhipaths, the corresponding path gains and path delays vary along with the mobile stations, and CCS methods fail to work in this dynamic scenario, A time-varying sparse channel estimate algorithm is proposed for MIMO OFDM systems, which is straightforwardly implemented in a stand-alone manner based on the well-known Kalman Filter （KF） formulation. Exact reconstruction is provided by converting the Re- weighed minimum el norm （RWel） problem into a KF with nonlinear equality constraints problem. In particularly, a Pseu- do-Measurement （PM） equation is formulated for the linearization of the RW^1 constrained equation. Thereafter, KF is employed twice to recover the time-varying sparse channel. Simulation results are presented to demonstrate that Mean Square Error （MSE） of the proposed algorithm is superior to the conventional algorithms, e.g. Basic Persuit （BP）, in the dynamic sparse scenario.

关 键 词：时变稀疏信道 多输入多输出正交频分复用系统 压缩感知 非线性等式约束卡尔曼 伪测量 

分 类 号：TN911.72[电子电信—通信与信息系统]