基于奇异值分解的稀疏信道估计

SPARSE CHANNEL ESTIMATION BASED ON SINGULAR VALUE DECOMPOSITION

针对基于压缩感知的信道估计能高效获取信道状态信息,以及噪声对估计算法的影响,提出一种基于奇异值分解的压缩感知估计算法。无需已知信道稀疏度,采用自适应步长使其重构精度和效率达到折中。引入奇异值分解技术,并根据奇异熵确定有效重构阶次,达到降噪目的,同时避免迭代过程中选取相关性较低的原子。仿真结果表明,该算法具有较高的重构精度,特别在低信噪比环境下,如信噪比为5 dB时,均方误差相对传统稀疏度自适应匹配追踪算法降低了95%左右,同时,算法运行时间也降低了约15%,具有较高的重构效率。

In view of efficiently obtain channel state information based on compressed sensing and the influence of noise on the estimation algorithm,this paper proposes a compressed sensing estimation algorithm based on singular value decomposition.It did not need to know the channel sparsity and used the adaptive step size to achieve a compromise between reconstruction accuracy and efficiency.The singular value decomposition technology was introduced,and the effective reconstruction order was determined according to the singular entropy to achieve the purpose of noise reduction.It avoided the selection of atoms with low correlation in the iterative process.The simulation results show that the algorithm has higher reconstruction accuracy,especially in low signal-to-noise ratio channel environment.For example,when the signal-to-noise ratio is 5 dB,the mean square error is reduced by about 95%,compared with the traditional Sparsity Adaptive Matching Pursuit(SAMP)algorithm.Meanwhile,the running time of the algorithm is reduced by about 15%,which has high reconstruction efficiency.