基于无网格压缩感知的DOA估计算法
发布时间：2018-12-12 01:10
【摘要】：应用传统的压缩感知理论对天线阵列信号的波达方向(Direction-of-arrival,DOA)进行估计,存在基的失配问题。基于交替方向乘子法(Alternative Direction Method of Multiplier,ADMM)的无网格压缩感知(Grid-less Compressive Sensing)技术能够解决该问题,但仍存在收敛速度慢的缺陷。针对该缺陷,提出带自适应惩罚项的ADMM(ADMM with adaptive penalty,AP-ADMM)算法,即根据输入信号的噪声功率,自适应地选择惩罚项的初始值;同时在算法迭代求解的过程中,自适应地对目标函数的惩罚项进行调整。与传统算法相比,在保证收敛精度和DOA的恢复成功概率的条件下,带自适应惩罚项的ADMM算法收敛速率明显加快。仿真结果验证了新算法的有效性。
[Abstract]:The traditional compression sensing theory is used to estimate the direction of arrival (Direction-of-arrival,DOA) of antenna array signals. Meshless compression sensing (Grid-less Compressive Sensing) based on alternating direction multiplier method (Alternative Direction Method of Multiplier,ADMM) can solve this problem, but it still has the defect of slow convergence. To solve this problem, an adaptive penalty term (ADMM (ADMM with adaptive penalty,AP-ADMM) algorithm is proposed, which adaptively selects the initial value of the penalty term according to the noise power of the input signal. At the same time, the penalty term of the objective function is adjusted adaptively in the iterative algorithm. Compared with the traditional algorithm, the convergence rate of the ADMM algorithm with adaptive penalty term is obviously accelerated under the condition of ensuring the convergence accuracy and the success probability of DOA recovery. Simulation results verify the effectiveness of the new algorithm.
【作者单位】： 中国人民解放军理工大学通信工程学院;
【基金】：国家自然科学基金(61571463,61371124,61272487,61472445,61201217)资助
【分类号】：TN911.7
，
本文编号：2373596
[Abstract]:The traditional compression sensing theory is used to estimate the direction of arrival (Direction-of-arrival,DOA) of antenna array signals. Meshless compression sensing (Grid-less Compressive Sensing) based on alternating direction multiplier method (Alternative Direction Method of Multiplier,ADMM) can solve this problem, but it still has the defect of slow convergence. To solve this problem, an adaptive penalty term (ADMM (ADMM with adaptive penalty,AP-ADMM) algorithm is proposed, which adaptively selects the initial value of the penalty term according to the noise power of the input signal. At the same time, the penalty term of the objective function is adjusted adaptively in the iterative algorithm. Compared with the traditional algorithm, the convergence rate of the ADMM algorithm with adaptive penalty term is obviously accelerated under the condition of ensuring the convergence accuracy and the success probability of DOA recovery. Simulation results verify the effectiveness of the new algorithm.
【作者单位】： 中国人民解放军理工大学通信工程学院;
【基金】：国家自然科学基金(61571463,61371124,61272487,61472445,61201217)资助
【分类号】：TN911.7
，
本文编号：2373596
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2373596.html
