基于FOCUSS二次加权的DOA估计方法
|
摘要
针对■-SVD、FOCUSS等稀疏重构算法应用波达方向(DOA)估计时,存在或运算量大、或精度不高的问题,提出了一种基于FOCUSS二次加权的信号DOA估计方法。将传统DOA估计表述为稀疏表示的信号模型,通过贝叶斯理论推导目标函数的最优解及加权矩阵,并在迭代过程中对结果进行二次加权优化,进一步增强恢复结果的稀疏性,提高恢复性能。仿真实验证明了该方法的优越性:与其他稀疏重构方法相比,该方法恢复精度高、稳健性好、运算量低。
When sparse reconstruction methods such as■-SVD and FOCUSS resolve the problem of DOA estimation,it causes problems of heavy computation,or low accuracy.To address this issue,the method of DOA estimation based on reweighted FOCUSS is proposed.Firstly,the traditional DOA estimation is represented as a sparse signal model.Then the optimal solution and weighted matrix are derived by applying Bayesian Theory.During each iteration,the estimation is further optimized through re-weighted proceeding,and it further enhances sparsity of recovery result to improve performance Simulation experiments are presented to prove the superiority of the method in this paper:comparing with other sparse reconstruction methods,the method in this paper possesses significant advantages of high recovery accuracy,good robustness and low computation.
When sparse reconstruction methods such as■-SVD and FOCUSS resolve the problem of DOA estimation,it causes problems of heavy computation,or low accuracy.To address this issue,the method of DOA estimation based on reweighted FOCUSS is proposed.Firstly,the traditional DOA estimation is represented as a sparse signal model.Then the optimal solution and weighted matrix are derived by applying Bayesian Theory.During each iteration,the estimation is further optimized through re-weighted proceeding,and it further enhances sparsity of recovery result to improve performance Simulation experiments are presented to prove the superiority of the method in this paper:comparing with other sparse reconstruction methods,the method in this paper possesses significant advantages of high recovery accuracy,good robustness and low computation.
引文
[1]韩学兵,杨建文.基于FOCUSS改进算法的DOA估计方法[J].火力与指挥控制,2014,39(7):93-96.
[2]PAL D,MENGSHOEL O.Stochastic CoSaMP:Randomizing greedy pursuit for sparse signal recovery[C]//Proc.ECMLPKDD,2016:761-776.
[3]MALIOUTOV D,CETIN M,WILLSKY A.A sparse signal reconstruction perspective for source localization with sensor arrays[J].IEEE Trans.Signal Process,2005,53(8):3010-3022.
[4]XIE K,HE Z,CICHOCKI A.Convergence analysis of the FOCUSS algorithm[J].IEEE Trans.Neural Learn Syst.,2015,26(3):601-613.
[5]HAN X,JIANG Z.RT-FOCUSS Algorithm for Sparse Recovery in Fully-perturbed Linear Model[C]//ACSR Advances in Comptuer Science Research,2015,15:573-576.
[6]韩学兵,张颢.模型噪声中的稀疏恢复算法研究[J].电子与信息学报,2012,34(8):1813-1818.
[7]CANDES E J.The restricted isometry property and its implications for compressed sensing[J].Academie des Sciences,Series I,2008,346(9-10):589-592.
[8]CANDES E J,WAKIN M B,BOYD S P,Enhancing Sparsity by reweighted minimization[J].J Fourier Anal Appl,2008,14:877-905.
[9]XU W,KHAJEHNEJAD M A,AVESTIMEHR A S,et al.Breaking through the Thresholds:an Analysis for Iterative Reweighted Minimization via the Grassmann Angle[C]//I-CASSP,2010:5498-5501.
[2]PAL D,MENGSHOEL O.Stochastic CoSaMP:Randomizing greedy pursuit for sparse signal recovery[C]//Proc.ECMLPKDD,2016:761-776.
[3]MALIOUTOV D,CETIN M,WILLSKY A.A sparse signal reconstruction perspective for source localization with sensor arrays[J].IEEE Trans.Signal Process,2005,53(8):3010-3022.
[4]XIE K,HE Z,CICHOCKI A.Convergence analysis of the FOCUSS algorithm[J].IEEE Trans.Neural Learn Syst.,2015,26(3):601-613.
[5]HAN X,JIANG Z.RT-FOCUSS Algorithm for Sparse Recovery in Fully-perturbed Linear Model[C]//ACSR Advances in Comptuer Science Research,2015,15:573-576.
[6]韩学兵,张颢.模型噪声中的稀疏恢复算法研究[J].电子与信息学报,2012,34(8):1813-1818.
[7]CANDES E J.The restricted isometry property and its implications for compressed sensing[J].Academie des Sciences,Series I,2008,346(9-10):589-592.
[8]CANDES E J,WAKIN M B,BOYD S P,Enhancing Sparsity by reweighted minimization[J].J Fourier Anal Appl,2008,14:877-905.
[9]XU W,KHAJEHNEJAD M A,AVESTIMEHR A S,et al.Breaking through the Thresholds:an Analysis for Iterative Reweighted Minimization via the Grassmann Angle[C]//I-CASSP,2010:5498-5501.