Nesterov快速迭代信道估计算法

A Fast Iterative Channel Estimation Algorithm Based on Nesterov Method

针对大规模多输入多输出(Multiple-Input Multiple-Output,MIMO)下行链路中,随着天线数的增加,信道估计的精度下降与复杂度大幅增加的问题,提出了一种基于交替迭代算法的快速信道估计算法。该算法利用大规模MIMO系统信道矩阵的稀疏特性,通过Nesterov平滑函数对压缩感知中的L1范数目标函数进行优化,并且每次迭代时通过构建目标函数以及交替迭代使算法收敛加速,在有效求解目标压缩感知问题的同时,方便求解目标函数的梯度,从而得到一种更加快速高效地求解L1范数最小化问题的算法。所提算法在高信噪比的条件下相较于传统的正交匹配追踪算法有很大的性能提升,同时算法收敛速度快于同类型的迭代算法。

To address the problem that the accuracy of channel estimation decreases as well as the complexity increases significantly with the increase of the number of antennas in the massive MIMO(Multiple-Input Multiple-Output) downlink, a fast channel estimation algorithm based on the alternating iterative algorithm is proposed. The algorithm takes advantage of the sparse features of the channel matrix of massive MIMO systems to optimize the L1-norm objective function in compressed sensing through the Nesterov smoothing function, and to speed up the convergence of the algorithm by constructing the objective function at each iteration and by alternating iterations. While solving the target compressed sensing problem effectively, this method can facilitate solving the gradient of the objective function, thus obtaining an algorithm for solving the L1-norm minimization problem more quickly and efficiently. Under the condition of high signal-tonoise ratio, the proposed algorithm has a great performance improvement compared with the conventional orthogonal matching pursuit algorithm, while the algorithm converges faster than the same type of iterative algorithm.