非整数倍路径时延下的OMP信道估计方法

OMP channel estimation method for non-integer multipath delay environment

正交匹配追踪(OMP)算法是一种基于压缩感知的稀疏信道估计方法,本文对OMP算法的终止条件进行研究,针对非整数倍路径时延环境中路径能量泄露导致路径总数增多且难以确定的问题,给出了一种基于噪声方差的终止准则,其中噪声方差可以通过估计获得。这种基于噪声方差的OMP算法(NV-OMP)能够检测信道路径个数,并获得有效的信道估计。仿真结果表明,与基于稀疏度已知的OMP算法相比,NV-OMP算法在非整数倍路径时延环境下的信道估计性能更加鲁棒,因而更适合于实际应用。文章同时仿真分析了噪声方差对NV-OMP算法性能的影响。

The Orthogonal Matching Pursuit（OMP） algorithm is one of the sparse channel estimation methods based on compressed sensing.Multipath energy leakage due to the non-integer multipath delay will always lead to multipath number increasing and bring difficulties to detection.In this paper,we investigate the termination criterion of the OMP algorithm and design one based on noise variance which can be obtained by estimation.The OMP algorithm based on noise variance（NV-OMP） can jointly detect the channel multipath number and estimate the channel state information.The stimulation results show that,compared with the OMP algorithm based on known sparsity,the NV-OMP algorithm is more robust in non-integer multipath delay environment,and hence is more suitable for practical applications.Moreover,the influence of noise variance estimation on the performance of the new OMP algorithm is also presented.